Final.ipynb

{
  "nbformat": 4,
  "nbformat_minor": 0,
  "metadata": {
    "colab": {
      "name": "Final.ipynb",
      "provenance": [],
      "collapsed_sections": [],
      "toc_visible": true,
      "authorship_tag": "ABX9TyPQHTsWLSOYvvPdGmHvLpqp"
    },
    "kernelspec": {
      "name": "python3",
      "display_name": "Python 3"
    }
  },
  "cells": [
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "SFKIbNtUjZdc"
      },
      "source": [
        "# COVID-19 in Academia"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "qM7ZsTbbe_vc"
      },
      "source": [
        "# Preparation"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "NyDEtvwlf4Fp"
      },
      "source": [
        "# install lucem_illud\n",
        "!pip install git+git://github.com/UChicago-Computational-Content-Analysis/lucem_illud.git"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "aHYhP-Qffy91"
      },
      "source": [
        "#All these packages need to be installed from pip\n",
        "import gensim#For word2vec, etc\n",
        "from gensim.models import ldaseqmodel\n",
        "import requests #For downloading our datasets\n",
        "import lucem_illud #pip install -U git+git://github.com/UChicago-Computational-Content-Analysis/lucem_illud.git\n",
        "\n",
        "import numpy as np #For arrays\n",
        "import pandas #Gives us DataFrames\n",
        "import matplotlib.pyplot as plt #For graphics\n",
        "import seaborn #Makes the graphics look nicer\n",
        "import sklearn.metrics.pairwise #For cosine similarity\n",
        "import sklearn.manifold #For T-SNE\n",
        "import sklearn.decomposition #For PCA\n",
        "import sklearn.cluster\n",
        "\n",
        "import matplotlib.pyplot as plt #For graphics\n",
        "import matplotlib.cm #Still for graphics\n",
        "\n",
        "#This 'magic' command makes the plots work better\n",
        "#in the notebook, don't use it outside of a notebook.\n",
        "#Also you can ignore the warning\n",
        "%matplotlib inline\n",
        "\n",
        "import os #For looking through files\n",
        "import os.path #For managing file paths"
      ],
      "execution_count": 5,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "F2l6RWTepJ_C"
      },
      "source": [
        "# import extra\n",
        "import gensim\n",
        "from textblob import TextBlob\n",
        "import scipy"
      ],
      "execution_count": 18,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "By0YwhNme-vd",
        "outputId": "383b6fef-74ec-438a-bf19-0dfcf8eab17e"
      },
      "source": [
        "# get papers\n",
        "temp_df = pandas.read_csv(\"/content/Papers_final.csv\")\n",
        "# delele NA years\n",
        "papers_df = temp_df.dropna(axis=0, how='any', subset=[\"year\"])\n",
        "# change year to int\n",
        "papers_df[\"year\"].astype(int)"
      ],
      "execution_count": 7,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "0       2021\n",
              "1       2021\n",
              "2       2021\n",
              "3       2021\n",
              "4       2021\n",
              "        ... \n",
              "1954    2019\n",
              "1955    2019\n",
              "1956    2019\n",
              "1957    2019\n",
              "1958    2019\n",
              "Name: year, Length: 1959, dtype: int64"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 7
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 1000
        },
        "id": "owBp-TaDe8iL",
        "outputId": "1488a627-b892-49f8-f046-fb6aa7111d38"
      },
      "source": [
        "# get econ data\n",
        "from collections import defaultdict\n",
        "from pathlib import Path\n",
        "\n",
        "my_dir_path = \"/content/econ\"\n",
        "\n",
        "results = defaultdict(list)\n",
        "for file in Path(my_dir_path).iterdir():\n",
        "    with open(file, \"r\") as file_open:\n",
        "        results[\"file_name\"].append(file.name)\n",
        "        results[\"text\"].append(file_open.read())\n",
        "econ_df = pandas.DataFrame(results)\n",
        "econ_df"
      ],
      "execution_count": 8,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>file_name</th>\n",
              "      <th>text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>019023-Polarization and Public Health- Partisa...</td>\n",
              "      <td>Polarization and Public Health: Partisan Diffe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>019059-Global Supply Chains in the Pandemic.txt</td>\n",
              "      <td>Global Supply Chains in the Pandemic*\\nBarthél...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>019037-ECONOMIC UNCERTAINTY BEFORE AND DURING ...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\nECONOMIC UNCERTAINT...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>019035-The Impact of COVID-19 on Student Exper...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE IMPACT OF COV...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>019051-How Did COVID-19 and Stabilization Poli...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\nHOW DID COVID-19 AN...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>019034-Fatalism, Beliefs, And Behaviors During...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nFATALISM, BELIEFS...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>019031-Aggregate and Firm-Level Stock Returns ...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nAGGREGATE AND FIR...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>019025-The Macroeconomics of Epidemics.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE MACROECONOMIC...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>019043-Early Evidence on the Impact of COVID-1...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nEARLY EVIDENCE ON...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>019032-An Seir Infectious Disease Model With T...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nAN SEIR INFECTIOU...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>019015-The Impact of COVID-19 on Gender Equali...</td>\n",
              "      <td>The Impact of COVID-19 on Gender Equality∗\\nTi...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>019049-Inequality in Household Adaptation to S...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nINEQUALITY IN HOU...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>019021-Optimal Mitigation Policies in a Pandem...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nOPTIMAL MITIGATIO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>019009-Policy Implications of Models of the Sp...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nPOLICY IMPLICATIO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>019042-The rise of COVID-19 cases is associate...</td>\n",
              "      <td>The rise of COVID-19 cases is associated with ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>019027-What Will be the Economic Impact of COV...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nWHAT WILL BE THE ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>019039-COVID-19 AND REMOTE WORK AN EARLY LOOK ...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCOVID-19 AND REMO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>17</th>\n",
              "      <td>019012-Determinants of Disparities in Covid-19...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nDETERMINANTS OF D...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>18</th>\n",
              "      <td>019019-A Simple Planning Problem for COVID-19 ...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nA SIMPLE PLANNING...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>19</th>\n",
              "      <td>019057-Implications of Heterogeneous SIR Model...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nIMPLICATIONS OF H...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>20</th>\n",
              "      <td>019001-Covid 19 Testing Inequality in New York...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCOVID-19: TESTING...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>21</th>\n",
              "      <td>019020-Health VS Wealth Public health policies...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHEALTH VS. WEALTH...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>22</th>\n",
              "      <td>019046-The Effectiveness of Life-Preserving In...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE EFFECTIVENESS...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>23</th>\n",
              "      <td>019011-Global Behaviors and Perceptions at the...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nGLOBAL BEHAVIORS ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>24</th>\n",
              "      <td>019026-Controlling a Pandemic.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCONTROLLING A PAN...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>25</th>\n",
              "      <td>019003-Human Mobility Restrictions and the Spr...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHUMAN MOBILITY RE...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>26</th>\n",
              "      <td>019018-Optimal targeted lockdowns in a multi-g...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nOPTIMAL TARGETED ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>27</th>\n",
              "      <td>019036-Voluntary and Mandatory Social Distanci...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nVOLUNTARY AND MAN...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>28</th>\n",
              "      <td>019008-COVID-Induced Economic Uncertainty.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCOVID-INDUCED ECO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>29</th>\n",
              "      <td>019024-Corporate Hiring under COVID-19.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCORPORATE HIRING ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>30</th>\n",
              "      <td>019029-Macroeconomic dynamics and reallocation...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nMACROECONOMIC DYN...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>31</th>\n",
              "      <td>019038-SOCIAL DISTANCING AND SOCIAL CAPITAL.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nSOCIAL DISTANCING...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>32</th>\n",
              "      <td>019005-The Geographic Spread of COVID-19 Corre...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE GEOGRAPHIC SP...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>33</th>\n",
              "      <td>019016-How does houselold spending respond to ...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHOW DOES HOUSEHOL...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>34</th>\n",
              "      <td>019033-Estimating the COVID-19 Infection Rate-...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nESTIMATING THE CO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>35</th>\n",
              "      <td>019030-Internal and External Effects of Social...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nINTERNAL AND EXTE...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>36</th>\n",
              "      <td>019052-Racial and Ethnic Disparities in COVID-...</td>\n",
              "      <td>Journal of Economics, Race, and Policy (2020) ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>37</th>\n",
              "      <td>019007-The Unprecedented Stock Market Impact o...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE UNPRECEDENTED...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>38</th>\n",
              "      <td>019006-How Many Jobs Can be Done at Home.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHOW MANY JOBS CAN...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>39</th>\n",
              "      <td>019048-How Much does COVID-19 Increase with Mo...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHOW MUCH DOES COV...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>40</th>\n",
              "      <td>019014-Ambiguous Pollution Response to COVID-1...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nAMBIGUOUS POLLUTI...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>41</th>\n",
              "      <td>019017-How valuable is financial flexibility w...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHOW VALUABLE IS F...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>42</th>\n",
              "      <td>019002-How Deadly Is COVID-19_ Understanding T...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nHOW DEADLY IS COV...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>43</th>\n",
              "      <td>019053-Macroeconomic Implications of COVID-19.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nMACROECONOMIC IMP...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>44</th>\n",
              "      <td>019013-The Adverse Effect of the COVID-19 Labo...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE ADVERSE EFFEC...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>45</th>\n",
              "      <td>019034-Sequential Lifting of COVID-19 Interven...</td>\n",
              "      <td>Sequential Lifting of COVID-19 Interventions w...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>46</th>\n",
              "      <td>019028-COVID-19 Infection Externalities.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCOVID-19 INFECTIO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>47</th>\n",
              "      <td>019045-COVID-19 and the Demand for Online Food...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nCOVID-19 AND THE ...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>48</th>\n",
              "      <td>019044-US Unemployment insurance replacement r...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nUS UNEMPLOYMENT I...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>49</th>\n",
              "      <td>019010-NBER, 2020 The Coronavirus and the Grea...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE CORONAVIRUS A...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>50</th>\n",
              "      <td>019033-The Cost of Privacy- Welfare Effect of ...</td>\n",
              "      <td>WORKING PAPER · NO. 2020-64\\n\\nThe Cost of Pri...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>51</th>\n",
              "      <td>019022-The Impacts of COVID-19 on Minority Une...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE IMPACTS OF CO...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>52</th>\n",
              "      <td>019050-Tracking Public and Private Responses t...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTRACKING PUBLIC A...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>53</th>\n",
              "      <td>019004-Labor Markets COVID-19.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nLABOR MARKETS DUR...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>54</th>\n",
              "      <td>019058-The Hammer and the Scalpel.txt</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE HAMMER AND TH...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                            file_name                                               text\n",
              "0   019023-Polarization and Public Health- Partisa...  Polarization and Public Health: Partisan Diffe...\n",
              "1     019059-Global Supply Chains in the Pandemic.txt  Global Supply Chains in the Pandemic*\\nBarthél...\n",
              "2   019037-ECONOMIC UNCERTAINTY BEFORE AND DURING ...  NBER WORKING PAPER SERIES\\nECONOMIC UNCERTAINT...\n",
              "3   019035-The Impact of COVID-19 on Student Exper...  NBER WORKING PAPER SERIES\\n\\nTHE IMPACT OF COV...\n",
              "4   019051-How Did COVID-19 and Stabilization Poli...  NBER WORKING PAPER SERIES\\nHOW DID COVID-19 AN...\n",
              "5   019034-Fatalism, Beliefs, And Behaviors During...  NBER WORKING PAPER SERIES\\n\\nFATALISM, BELIEFS...\n",
              "6   019031-Aggregate and Firm-Level Stock Returns ...  NBER WORKING PAPER SERIES\\n\\nAGGREGATE AND FIR...\n",
              "7          019025-The Macroeconomics of Epidemics.txt  NBER WORKING PAPER SERIES\\n\\nTHE MACROECONOMIC...\n",
              "8   019043-Early Evidence on the Impact of COVID-1...  NBER WORKING PAPER SERIES\\n\\nEARLY EVIDENCE ON...\n",
              "9   019032-An Seir Infectious Disease Model With T...  NBER WORKING PAPER SERIES\\n\\nAN SEIR INFECTIOU...\n",
              "10  019015-The Impact of COVID-19 on Gender Equali...  The Impact of COVID-19 on Gender Equality∗\\nTi...\n",
              "11  019049-Inequality in Household Adaptation to S...  NBER WORKING PAPER SERIES\\n\\nINEQUALITY IN HOU...\n",
              "12  019021-Optimal Mitigation Policies in a Pandem...  NBER WORKING PAPER SERIES\\n\\nOPTIMAL MITIGATIO...\n",
              "13  019009-Policy Implications of Models of the Sp...  NBER WORKING PAPER SERIES\\n\\nPOLICY IMPLICATIO...\n",
              "14  019042-The rise of COVID-19 cases is associate...  The rise of COVID-19 cases is associated with ...\n",
              "15  019027-What Will be the Economic Impact of COV...  NBER WORKING PAPER SERIES\\n\\nWHAT WILL BE THE ...\n",
              "16  019039-COVID-19 AND REMOTE WORK AN EARLY LOOK ...  NBER WORKING PAPER SERIES\\n\\nCOVID-19 AND REMO...\n",
              "17  019012-Determinants of Disparities in Covid-19...  NBER WORKING PAPER SERIES\\n\\nDETERMINANTS OF D...\n",
              "18  019019-A Simple Planning Problem for COVID-19 ...  NBER WORKING PAPER SERIES\\n\\nA SIMPLE PLANNING...\n",
              "19  019057-Implications of Heterogeneous SIR Model...  NBER WORKING PAPER SERIES\\n\\nIMPLICATIONS OF H...\n",
              "20  019001-Covid 19 Testing Inequality in New York...  NBER WORKING PAPER SERIES\\n\\nCOVID-19: TESTING...\n",
              "21  019020-Health VS Wealth Public health policies...  NBER WORKING PAPER SERIES\\n\\nHEALTH VS. WEALTH...\n",
              "22  019046-The Effectiveness of Life-Preserving In...  NBER WORKING PAPER SERIES\\n\\nTHE EFFECTIVENESS...\n",
              "23  019011-Global Behaviors and Perceptions at the...  NBER WORKING PAPER SERIES\\n\\nGLOBAL BEHAVIORS ...\n",
              "24                  019026-Controlling a Pandemic.txt  NBER WORKING PAPER SERIES\\n\\nCONTROLLING A PAN...\n",
              "25  019003-Human Mobility Restrictions and the Spr...  NBER WORKING PAPER SERIES\\n\\nHUMAN MOBILITY RE...\n",
              "26  019018-Optimal targeted lockdowns in a multi-g...  NBER WORKING PAPER SERIES\\n\\nOPTIMAL TARGETED ...\n",
              "27  019036-Voluntary and Mandatory Social Distanci...  NBER WORKING PAPER SERIES\\n\\nVOLUNTARY AND MAN...\n",
              "28      019008-COVID-Induced Economic Uncertainty.txt  NBER WORKING PAPER SERIES\\n\\nCOVID-INDUCED ECO...\n",
              "29         019024-Corporate Hiring under COVID-19.txt  NBER WORKING PAPER SERIES\\n\\nCORPORATE HIRING ...\n",
              "30  019029-Macroeconomic dynamics and reallocation...  NBER WORKING PAPER SERIES\\n\\nMACROECONOMIC DYN...\n",
              "31    019038-SOCIAL DISTANCING AND SOCIAL CAPITAL.txt  NBER WORKING PAPER SERIES\\n\\nSOCIAL DISTANCING...\n",
              "32  019005-The Geographic Spread of COVID-19 Corre...  NBER WORKING PAPER SERIES\\n\\nTHE GEOGRAPHIC SP...\n",
              "33  019016-How does houselold spending respond to ...  NBER WORKING PAPER SERIES\\n\\nHOW DOES HOUSEHOL...\n",
              "34  019033-Estimating the COVID-19 Infection Rate-...  NBER WORKING PAPER SERIES\\n\\nESTIMATING THE CO...\n",
              "35  019030-Internal and External Effects of Social...  NBER WORKING PAPER SERIES\\n\\nINTERNAL AND EXTE...\n",
              "36  019052-Racial and Ethnic Disparities in COVID-...  Journal of Economics, Race, and Policy (2020) ...\n",
              "37  019007-The Unprecedented Stock Market Impact o...  NBER WORKING PAPER SERIES\\n\\nTHE UNPRECEDENTED...\n",
              "38       019006-How Many Jobs Can be Done at Home.txt  NBER WORKING PAPER SERIES\\n\\nHOW MANY JOBS CAN...\n",
              "39  019048-How Much does COVID-19 Increase with Mo...  NBER WORKING PAPER SERIES\\n\\nHOW MUCH DOES COV...\n",
              "40  019014-Ambiguous Pollution Response to COVID-1...  NBER WORKING PAPER SERIES\\n\\nAMBIGUOUS POLLUTI...\n",
              "41  019017-How valuable is financial flexibility w...  NBER WORKING PAPER SERIES\\n\\nHOW VALUABLE IS F...\n",
              "42  019002-How Deadly Is COVID-19_ Understanding T...  NBER WORKING PAPER SERIES\\n\\nHOW DEADLY IS COV...\n",
              "43  019053-Macroeconomic Implications of COVID-19.txt  NBER WORKING PAPER SERIES\\n\\nMACROECONOMIC IMP...\n",
              "44  019013-The Adverse Effect of the COVID-19 Labo...  NBER WORKING PAPER SERIES\\n\\nTHE ADVERSE EFFEC...\n",
              "45  019034-Sequential Lifting of COVID-19 Interven...  Sequential Lifting of COVID-19 Interventions w...\n",
              "46        019028-COVID-19 Infection Externalities.txt  NBER WORKING PAPER SERIES\\n\\nCOVID-19 INFECTIO...\n",
              "47  019045-COVID-19 and the Demand for Online Food...  NBER WORKING PAPER SERIES\\n\\nCOVID-19 AND THE ...\n",
              "48  019044-US Unemployment insurance replacement r...  NBER WORKING PAPER SERIES\\n\\nUS UNEMPLOYMENT I...\n",
              "49  019010-NBER, 2020 The Coronavirus and the Grea...  NBER WORKING PAPER SERIES\\n\\nTHE CORONAVIRUS A...\n",
              "50  019033-The Cost of Privacy- Welfare Effect of ...  WORKING PAPER · NO. 2020-64\\n\\nThe Cost of Pri...\n",
              "51  019022-The Impacts of COVID-19 on Minority Une...  NBER WORKING PAPER SERIES\\n\\nTHE IMPACTS OF CO...\n",
              "52  019050-Tracking Public and Private Responses t...  NBER WORKING PAPER SERIES\\n\\nTRACKING PUBLIC A...\n",
              "53                  019004-Labor Markets COVID-19.txt  NBER WORKING PAPER SERIES\\n\\nLABOR MARKETS DUR...\n",
              "54              019058-The Hammer and the Scalpel.txt  NBER WORKING PAPER SERIES\\n\\nTHE HAMMER AND TH..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 8
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 570
        },
        "id": "51Ye1VldhqqI",
        "outputId": "a067203e-215e-4a24-cd02-7bf0fcd7d30c"
      },
      "source": [
        "# get psych data\n",
        "from collections import defaultdict\n",
        "from pathlib import Path\n",
        "\n",
        "my_dir_path = \"/content/psych\"\n",
        "\n",
        "results = defaultdict(list)\n",
        "for file in Path(my_dir_path).iterdir():\n",
        "    with open(file, \"r\") as file_open:\n",
        "        results[\"file_name\"].append(file.name)\n",
        "        results[\"text\"].append(file_open.read())\n",
        "psych_df = pandas.DataFrame(results)\n",
        "psych_df"
      ],
      "execution_count": 9,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>file_name</th>\n",
              "      <th>text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>Changes to the Work–Family Interface During th...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>Economic Stressors and the Enactment of CDC-Re...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>Working in a Pandemic.txt</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>Anxiety Responses to the Unfolding COVID-19 Cr...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>When There is a Will There is a Way.txt</td>\n",
              "      <td>Journal of Applied Psychology\\nWhen There is a...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>The Mind, the Heart, and the Leader in Times o...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>Supporting Interdependent Telework Employees.txt</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>How Critical Activities Within COVID-19 Intens...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>Tackling the Negative Impact of COVID-19 on Wo...</td>\n",
              "      <td>Journal of Applied Psychology\\nTackling the Ne...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>Socioeconomic Status and Well-Being During COV...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>10</th>\n",
              "      <td>From Zero to Hero.txt</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>11</th>\n",
              "      <td>Autonomy Restoration During a Global Pandemic.txt</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>When Helping Hurts COVID-19 Critical Incident ...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>Impact of the COVID-19 Pandemic on Job Search ...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>Plug Back Into Work, Safely Job Reattachment, ...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>Women’s Leadership Is Associated With Fewer De...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>Work-Family Strategies During COVID-19 Examini...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "                                            file_name                                               text\n",
              "0   Changes to the Work–Family Interface During th...  Journal of Applied Psychology\\n© 2020 American...\n",
              "1   Economic Stressors and the Enactment of CDC-Re...  Journal of Applied Psychology\\n© 2020 American...\n",
              "2                           Working in a Pandemic.txt  Journal of Applied Psychology\\n© 2020 American...\n",
              "3   Anxiety Responses to the Unfolding COVID-19 Cr...  Journal of Applied Psychology\\n© 2020 American...\n",
              "4             When There is a Will There is a Way.txt  Journal of Applied Psychology\\nWhen There is a...\n",
              "5   The Mind, the Heart, and the Leader in Times o...  Journal of Applied Psychology\\n© 2020 American...\n",
              "6    Supporting Interdependent Telework Employees.txt  Journal of Applied Psychology\\n© 2020 American...\n",
              "7   How Critical Activities Within COVID-19 Intens...  Journal of Applied Psychology\\n© 2020 American...\n",
              "8   Tackling the Negative Impact of COVID-19 on Wo...  Journal of Applied Psychology\\nTackling the Ne...\n",
              "9   Socioeconomic Status and Well-Being During COV...  Journal of Applied Psychology\\n© 2020 American...\n",
              "10                              From Zero to Hero.txt  Journal of Applied Psychology\\n© 2020 American...\n",
              "11  Autonomy Restoration During a Global Pandemic.txt  Journal of Applied Psychology\\n© 2020 American...\n",
              "12  When Helping Hurts COVID-19 Critical Incident ...  Journal of Applied Psychology\\n© 2020 American...\n",
              "13  Impact of the COVID-19 Pandemic on Job Search ...  Journal of Applied Psychology\\n© 2020 American...\n",
              "14  Plug Back Into Work, Safely Job Reattachment, ...  Journal of Applied Psychology\\n© 2020 American...\n",
              "15  Women’s Leadership Is Associated With Fewer De...  Journal of Applied Psychology\\n© 2020 American...\n",
              "16  Work-Family Strategies During COVID-19 Examini...  Journal of Applied Psychology\\n© 2020 American..."
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 9
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 415
        },
        "id": "DShWrVwli64y",
        "outputId": "487906b5-282e-42cf-e61d-f15dc950a70a"
      },
      "source": [
        "# combine econ and psych\n",
        "full_df = pandas.concat([econ_df,psych_df])\n",
        "full_df"
      ],
      "execution_count": 10,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>file_name</th>\n",
              "      <th>text</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>019023-Polarization and Public Health- Partisa...</td>\n",
              "      <td>Polarization and Public Health: Partisan Diffe...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>019059-Global Supply Chains in the Pandemic.txt</td>\n",
              "      <td>Global Supply Chains in the Pandemic*\\nBarthél...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>019037-ECONOMIC UNCERTAINTY BEFORE AND DURING ...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\nECONOMIC UNCERTAINT...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>019035-The Impact of COVID-19 on Student Exper...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\n\\nTHE IMPACT OF COV...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>019051-How Did COVID-19 and Stabilization Poli...</td>\n",
              "      <td>NBER WORKING PAPER SERIES\\nHOW DID COVID-19 AN...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>...</th>\n",
              "      <td>...</td>\n",
              "      <td>...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>12</th>\n",
              "      <td>When Helping Hurts COVID-19 Critical Incident ...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>13</th>\n",
              "      <td>Impact of the COVID-19 Pandemic on Job Search ...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>14</th>\n",
              "      <td>Plug Back Into Work, Safely Job Reattachment, ...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>15</th>\n",
              "      <td>Women’s Leadership Is Associated With Fewer De...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>16</th>\n",
              "      <td>Work-Family Strategies During COVID-19 Examini...</td>\n",
              "      <td>Journal of Applied Psychology\\n© 2020 American...</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "<p>72 rows × 2 columns</p>\n",
              "</div>"
            ],
            "text/plain": [
              "                                            file_name                                               text\n",
              "0   019023-Polarization and Public Health- Partisa...  Polarization and Public Health: Partisan Diffe...\n",
              "1     019059-Global Supply Chains in the Pandemic.txt  Global Supply Chains in the Pandemic*\\nBarthél...\n",
              "2   019037-ECONOMIC UNCERTAINTY BEFORE AND DURING ...  NBER WORKING PAPER SERIES\\nECONOMIC UNCERTAINT...\n",
              "3   019035-The Impact of COVID-19 on Student Exper...  NBER WORKING PAPER SERIES\\n\\nTHE IMPACT OF COV...\n",
              "4   019051-How Did COVID-19 and Stabilization Poli...  NBER WORKING PAPER SERIES\\nHOW DID COVID-19 AN...\n",
              "..                                                ...                                                ...\n",
              "12  When Helping Hurts COVID-19 Critical Incident ...  Journal of Applied Psychology\\n© 2020 American...\n",
              "13  Impact of the COVID-19 Pandemic on Job Search ...  Journal of Applied Psychology\\n© 2020 American...\n",
              "14  Plug Back Into Work, Safely Job Reattachment, ...  Journal of Applied Psychology\\n© 2020 American...\n",
              "15  Women’s Leadership Is Associated With Fewer De...  Journal of Applied Psychology\\n© 2020 American...\n",
              "16  Work-Family Strategies During COVID-19 Examini...  Journal of Applied Psychology\\n© 2020 American...\n",
              "\n",
              "[72 rows x 2 columns]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 10
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "H8kpss8siBzs"
      },
      "source": [
        "# tokenization\n",
        "full_df['tokenized_sents'] = full_df['text'].apply(lambda x: [lucem_illud.word_tokenize(s) for s in lucem_illud.sent_tokenize(x)])\n",
        "# lemma\n",
        "full_df['normalized_sents'] = full_df['tokenized_sents'].apply(lambda x: [lucem_illud.normalizeTokens(s, lemma=True) for s in x])\n",
        "\n",
        "full_df['tokenized_text'] = full_df['text'].apply(lambda x: lucem_illud.word_tokenize(x))\n",
        "full_df['normalized_tokens'] = full_df['tokenized_text'].apply(lambda x: lucem_illud.normalizeTokens(x))"
      ],
      "execution_count": 11,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "m51gVeGecXiN"
      },
      "source": [
        "full_df"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Bsl0UaG3kUHP"
      },
      "source": [
        "# tokenization\n",
        "papers_df['tokenized_sents'] = papers_df['text'].apply(lambda x: [lucem_illud.word_tokenize(s) for s in lucem_illud.sent_tokenize(x)])\n",
        "# lemma\n",
        "papers_df['normalized_sents'] = papers_df['tokenized_sents'].apply(lambda x: [lucem_illud.normalizeTokens(s, lemma=True) for s in x])\n",
        "\n",
        "papers_df['tokenized_text'] = papers_df['text'].apply(lambda x: lucem_illud.word_tokenize(x))\n",
        "papers_df['normalized_tokens'] = papers_df['tokenized_text'].apply(lambda x: lucem_illud.normalizeTokens(x))"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "YYcR4bdpkewv"
      },
      "source": [
        "papers_df"
      ],
      "execution_count": null,
      "outputs": []
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "9MJC5K1JkNFg"
      },
      "source": [
        "# KL Divergence\n"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "U6_nhC7dkegm",
        "outputId": "1b1bca79-e004-4eb6-8156-c7329b5c3373"
      },
      "source": [
        "# remove overlap\n",
        "fullWords = set(full_df['normalized_tokens'].sum())\n",
        "papersWords = set(papers_df['normalized_tokens'].sum())\n",
        "\n",
        "#Change & to | if you want to keep all words\n",
        "overlapWords = fullWords & papersWords\n",
        "\n",
        "overlapWordsDict = {word: index for index, word in enumerate(overlapWords)}\n",
        "overlapWordsDict['work']"
      ],
      "execution_count": 13,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "666"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 13
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "GheFLnwUo2Nz"
      },
      "source": [
        "def makeProbsArray(dfColumn, overlapDict):\n",
        "    words = dfColumn.sum()\n",
        "    countList = [0] * len(overlapDict)\n",
        "    for word in words:\n",
        "        try:\n",
        "            countList[overlapDict[word]] += 1\n",
        "        except KeyError:\n",
        "            #The word is not common so we skip it\n",
        "            pass\n",
        "    countArray = np.array(countList)\n",
        "    return countArray / countArray.sum()"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "iut8a6cwosI2",
        "outputId": "b05d46c6-bf06-4e05-bc5c-1499ab8ef26f"
      },
      "source": [
        "fullProbArray = makeProbsArray(full_df['normalized_tokens'], overlapWordsDict)\n",
        "papersProbArray = makeProbsArray(papers_df['normalized_tokens'], overlapWordsDict)\n",
        "papersProbArray.sum()"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "1.0"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vOk3_yQco4BV",
        "outputId": "b1402926-c080-4fc6-c165-3740fa4cc501"
      },
      "source": [
        "# KL convergence\n",
        "social_naturalDivergence = scipy.stats.entropy(fullProbArray, papersProbArray)\n",
        "print (social_naturalDivergence)\n",
        "natural_socialDivergence = scipy.stats.entropy(papersProbArray, fullProbArray)\n",
        "print (natural_socialDivergence)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "1.0003624288133506\n",
            "1.0638245866284137\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "-eJXtZVpjfWF"
      },
      "source": [
        "# Word2Vec"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "CPu1LXMeZ0AL"
      },
      "source": [
        "## Social Science"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "O3yDlwIHkO28"
      },
      "source": [
        "fullW2V = gensim.models.word2vec.Word2Vec(full_df['normalized_sents'].sum())"
      ],
      "execution_count": 13,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "KGEK1yLF8_W6",
        "outputId": "7af9c9f2-c741-4556-a7c0-d663847084b7"
      },
      "source": [
        "fullW2V.most_similar('president')"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('november', 0.9819780588150024),\n",
              " ('september', 0.9787055253982544),\n",
              " ('arkansas', 0.9778821468353271),\n",
              " ('la', 0.976869523525238),\n",
              " ('october', 0.9767385125160217),\n",
              " ('hawaii', 0.9765076637268066),\n",
              " ('24th', 0.9749261140823364),\n",
              " ('louisiana', 0.9748409986495972),\n",
              " ('gretchen', 0.9747686386108398),\n",
              " ('carolina', 0.9740902185440063)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "c1yvCcFmkVSm",
        "outputId": "806dbfd4-91aa-41db-e4a2-4cd08aa44b52"
      },
      "source": [
        "fullW2V.most_similar('covid-19')"
      ],
      "execution_count": 14,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('outbreak', 0.849372386932373),\n",
              " ('influenza', 0.8349153995513916),\n",
              " ('pandemic', 0.8320005536079407),\n",
              " ('corona', 0.8232523202896118),\n",
              " ('covid19', 0.8178908824920654),\n",
              " ('spanish', 0.811830997467041),\n",
              " ('coronavirus', 0.8107658624649048),\n",
              " ('response', 0.80597984790802),\n",
              " ('novel', 0.8049942255020142),\n",
              " ('covid', 0.800186038017273)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 14
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5-pWb4XqkYvC",
        "outputId": "9d11995d-ebca-455b-b9af-bd554b479129"
      },
      "source": [
        "fullW2V.most_similar('vaccine')"
      ],
      "execution_count": 15,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('infectiousness', 0.9439640045166016),\n",
              " ('effective', 0.9404677152633667),\n",
              " ('cure', 0.9244608283042908),\n",
              " ('know', 0.9193854331970215),\n",
              " ('extreme', 0.9165095686912537),\n",
              " ('quarantine', 0.9129801392555237),\n",
              " ('isolation', 0.9125193357467651),\n",
              " ('trace', 0.9110430479049683),\n",
              " ('lockdowns', 0.9108986258506775),\n",
              " ('contract', 0.9098805785179138)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 15
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "EPfhThxukc9v",
        "outputId": "2fad9cc1-3cbf-4dfd-b3bd-048581093d0d"
      },
      "source": [
        "fullW2V.most_similar('treatment')"
      ],
      "execution_count": 16,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('belief', 0.9609736800193787),\n",
              " ('vary', 0.9371767640113831),\n",
              " ('differential', 0.9222626686096191),\n",
              " ('interpret', 0.9134619235992432),\n",
              " ('additional', 0.9041493535041809),\n",
              " ('differ', 0.9014716148376465),\n",
              " ('overall', 0.8998055458068848),\n",
              " ('scenario', 0.8960479497909546),\n",
              " ('hold', 0.8951650857925415),\n",
              " ('trajectory', 0.8904632329940796)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 16
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ThbVay7SkiK1",
        "outputId": "a860d72c-8199-422d-c9ec-77395e82f2a0"
      },
      "source": [
        "fullW2V.most_similar('china')"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('epicenter', 0.9148597717285156),\n",
              " ('2019-ncov', 0.9146484136581421),\n",
              " ('italy', 0.9090564250946045),\n",
              " ('outbreak', 0.908423900604248),\n",
              " ('coivd-19', 0.8972347974777222),\n",
              " ('corona', 0.8589129447937012),\n",
              " ('wuhan', 0.8528329730033875),\n",
              " ('worldwide', 0.8503612279891968),\n",
              " ('coronavirus', 0.8499854803085327),\n",
              " ('novel', 0.8499071002006531)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "0pMxMzmXkqoR",
        "outputId": "9df7900a-5e0b-4c43-98da-3a7a8dc227a5"
      },
      "source": [
        "fullW2V.most_similar('america')"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('warn', 0.9778646230697632),\n",
              " ('september', 0.9714869856834412),\n",
              " ('japan', 0.9712555408477783),\n",
              " ('france', 0.9693090915679932),\n",
              " ('germany', 0.9680418372154236),\n",
              " ('oklahoma', 0.9666305780410767),\n",
              " ('6', 0.9656875133514404),\n",
              " ('president', 0.9641669988632202),\n",
              " ('alabama', 0.9641361236572266),\n",
              " ('russia', 0.9635642766952515)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FyCT7N8Okumy",
        "outputId": "f3009d68-a2a2-4f6e-c99e-c5b5ddbf3ec6"
      },
      "source": [
        "fullW2V.most_similar('state')"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('kingdom', 0.8280285596847534),\n",
              " ('unite', 0.7875370979309082),\n",
              " ('county', 0.7582467794418335),\n",
              " ('reopen', 0.7544432878494263),\n",
              " ('governor', 0.7512765526771545),\n",
              " ('steady', 0.748540997505188),\n",
              " ('nation', 0.7378343343734741),\n",
              " ('country', 0.7249220013618469),\n",
              " ('declaration', 0.7214992046356201),\n",
              " ('resuscitation', 0.7055197954177856)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "s7ZWFYKKlJsD",
        "outputId": "3316c611-b5c4-4211-83d0-caa7691d0d32"
      },
      "source": [
        "fullW2V.most_similar('people')"
      ],
      "execution_count": 23,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('require', 0.8840236067771912),\n",
              " ('individual', 0.8774884939193726),\n",
              " ('avoid', 0.8753497004508972),\n",
              " ('get', 0.8668035864830017),\n",
              " ('isolate', 0.8601580858230591),\n",
              " ('infect', 0.8601499795913696),\n",
              " ('meet', 0.8599317073822021),\n",
              " ('sick', 0.8542327284812927),\n",
              " ('know', 0.8538702130317688),\n",
              " ('asymptomatic', 0.8528835773468018)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 23
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "SQteQm-wlXgN",
        "outputId": "79367d9a-61e9-4e13-de40-01378f00aafe"
      },
      "source": [
        "fullW2V.most_similar('worker')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('child', 0.9048381447792053),\n",
              " ('member', 0.8561595678329468),\n",
              " ('employ', 0.8439839482307434),\n",
              " ('household', 0.8243786692619324),\n",
              " ('live', 0.8202165365219116),\n",
              " ('likely', 0.8187476396560669),\n",
              " ('essential', 0.7989829778671265),\n",
              " ('sale', 0.7958325743675232),\n",
              " ('couple', 0.7945008277893066),\n",
              " ('lose', 0.7865341901779175)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "xNXLPlIeGNrm",
        "outputId": "22dc8ed1-5c77-45ff-d665-9b8ed346d978"
      },
      "source": [
        "fullW2V.most_similar('job')"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('performance', 0.7767320871353149),\n",
              " ('perform', 0.7687323093414307),\n",
              " ('task', 0.7589665055274963),\n",
              " ('student', 0.7334562540054321),\n",
              " ('experience', 0.7280638217926025),\n",
              " ('insecurity', 0.7277714014053345),\n",
              " ('couple', 0.7221989631652832),\n",
              " ('woman', 0.7134095430374146),\n",
              " ('skill', 0.7128334045410156),\n",
              " ('workplace', 0.7101960778236389)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "voNQkp9YlaGV",
        "outputId": "43620d20-54e1-4237-f5ed-bb8b789b6e9f"
      },
      "source": [
        "fullW2V.most_similar('leader')"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('commitment', 0.9728037714958191),\n",
              " ('indirectly', 0.9477722644805908),\n",
              " ('knowledge', 0.944723904132843),\n",
              " ('safety', 0.9422106742858887),\n",
              " ('contribute', 0.9403423070907593),\n",
              " ('exacerbate', 0.9397084712982178),\n",
              " ('progress', 0.9394020438194275),\n",
              " ('goal', 0.9372594356536865),\n",
              " ('regulation', 0.9357172846794128),\n",
              " ('potential', 0.9339116811752319)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "g2CkMMsble1u",
        "outputId": "c3dae2ad-85fa-421c-8232-fb28b671023c"
      },
      "source": [
        "fullW2V.most_similar('psychological')"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('association', 0.9376375079154968),\n",
              " ('document', 0.9167454838752747),\n",
              " ('american', 0.8909935355186462),\n",
              " ('ally', 0.8893635272979736),\n",
              " ('psychologist', 0.8778862953186035),\n",
              " ('fulfillment', 0.8642103672027588),\n",
              " ('copyright', 0.8618373870849609),\n",
              " ('publisher', 0.8573639392852783),\n",
              " ('perspective', 0.8492352962493896),\n",
              " ('513–524', 0.8421583771705627)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "r6OBI15Clhr3",
        "outputId": "8e9eeec0-0fd6-456e-9297-b2399b31524e"
      },
      "source": [
        "fullW2V.most_similar('economy')"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('shock', 0.9052436351776123),\n",
              " ('chain', 0.898902177810669),\n",
              " ('pharmaceutical', 0.8619531393051147),\n",
              " ('mitigate', 0.8553270101547241),\n",
              " ('externality', 0.8548004627227783),\n",
              " ('equilibrium', 0.8537750244140625),\n",
              " ('temporary', 0.8504856824874878),\n",
              " ('trade', 0.8500582575798035),\n",
              " ('consequence', 0.8491602540016174),\n",
              " ('affect', 0.8470218777656555)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "q3rJi5KimKGh",
        "outputId": "46df8252-9d0d-4341-fa82-460f65544106"
      },
      "source": [
        "numWords = 50\n",
        "targetWords = fullW2V.wv.index2word[:numWords]\n",
        "\n",
        "wordsSubMatrix = []\n",
        "for word in targetWords:\n",
        "    wordsSubMatrix.append(fullW2V[word])\n",
        "wordsSubMatrix = np.array(wordsSubMatrix)\n",
        "wordsSubMatrix"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[ 0.5597746 ,  0.11376412, -0.46149015, ..., -1.0161502 ,\n",
              "        -0.15520696,  1.9305593 ],\n",
              "       [ 1.5558631 ,  0.6378739 , -0.28709278, ...,  0.01625011,\n",
              "        -0.33888507, -0.41665804],\n",
              "       [ 0.13115358, -0.79369175, -0.08601692, ...,  0.47851998,\n",
              "         1.3679327 ,  0.29697704],\n",
              "       ...,\n",
              "       [ 1.0241284 ,  0.59538925, -0.52049774, ..., -0.9568455 ,\n",
              "         0.6599825 ,  1.6414175 ],\n",
              "       [ 0.8711913 ,  0.400333  , -0.61866343, ..., -0.33490178,\n",
              "         1.0486008 , -0.33106804],\n",
              "       [ 0.82725126, -0.5059644 , -0.27818385, ...,  0.06999511,\n",
              "         0.33809572, -0.779389  ]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0ZAgIUVImR2w"
      },
      "source": [
        "pcaWords = sklearn.decomposition.PCA(n_components = 50).fit(wordsSubMatrix)\n",
        "reducedPCA_data = pcaWords.transform(wordsSubMatrix)\n",
        "# TSNE\n",
        "tsneWords = sklearn.manifold.TSNE(n_components = 2).fit_transform(reducedPCA_data)"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 935
        },
        "id": "LGpqXq0Bmasc",
        "outputId": "56136823-f1c5-4595-ee10-8727712b5989"
      },
      "source": [
        "fig = plt.figure(figsize = (15,15))\n",
        "ax = fig.add_subplot(111)\n",
        "ax.set_frame_on(False)\n",
        "plt.scatter(tsneWords[:, 0], tsneWords[:, 1], alpha = 0)#Making the points invisible \n",
        "for i, word in enumerate(targetWords):\n",
        "    ax.annotate(word, (tsneWords[:, 0][i],tsneWords[:, 1][i]), size =  20 * (numWords - i) / numWords)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/matplotlib/backends/backend_agg.py:214: RuntimeWarning: Glyph 11005 missing from current font.\n",
            "  font.set_text(s, 0.0, flags=flags)\n",
            "/usr/local/lib/python3.7/dist-packages/matplotlib/backends/backend_agg.py:183: RuntimeWarning: Glyph 11005 missing from current font.\n",
            "  font.set_text(s, 0, flags=flags)\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x1080 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Dms-aBUbk4Zf"
      },
      "source": [
        "## Natural Science"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Si6oxofSk6q9"
      },
      "source": [
        "papersW2V = gensim.models.word2vec.Word2Vec(papers_df['normalized_sents'].sum())"
      ],
      "execution_count": 15,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "wn2L1jRNlZdp",
        "outputId": "13adf862-2146-4e62-c2a5-5d0e17236e72"
      },
      "source": [
        "papersW2V.most_similar('covid-19')"
      ],
      "execution_count": 17,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('new', 0.9998427629470825),\n",
              " ('result', 0.9998418092727661),\n",
              " ('continue', 0.9998132586479187),\n",
              " ('human', 0.9997997879981995),\n",
              " ('recently', 0.9997830390930176),\n",
              " ('complication', 0.9997658729553223),\n",
              " ('late', 0.9997572898864746),\n",
              " ('infectious', 0.9997531771659851),\n",
              " ('worldwide', 0.9997470378875732),\n",
              " ('begin', 0.9997299313545227)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 17
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "NprAuviImD16",
        "outputId": "866ce92a-ce58-4d60-b825-2098c2ef0901"
      },
      "source": [
        "papersW2V.most_similar('vaccine')"
      ],
      "execution_count": 21,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('model', 0.9998939037322998),\n",
              " ('effective', 0.9998881816864014),\n",
              " ('potential', 0.9998874664306641),\n",
              " ('strategy', 0.999883234500885),\n",
              " ('include', 0.9998825788497925),\n",
              " ('drug', 0.9998806715011597),\n",
              " ('understand', 0.9998775720596313),\n",
              " ('early', 0.9998756647109985),\n",
              " ('reduce', 0.9998752474784851),\n",
              " ('control', 0.9998699426651001)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 21
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "dGqyUPH2mGb4",
        "outputId": "8cd6140c-2778-4aa6-cdfd-6d72d5724e0c"
      },
      "source": [
        "papersW2V.most_similar('treatment')"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('low', 0.9998729228973389),\n",
              " ('relate', 0.9998582601547241),\n",
              " ('understand', 0.9998553395271301),\n",
              " ('vaccine', 0.9998540878295898),\n",
              " ('patient', 0.9998528957366943),\n",
              " ('present', 0.999847412109375),\n",
              " ('severity', 0.9998459815979004),\n",
              " ('find', 0.9998447895050049),\n",
              " ('trial', 0.9998437166213989),\n",
              " ('march', 0.9998431205749512)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 22
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Luqhx4iXmIvY",
        "outputId": "c7995c6a-ed5b-4afd-b0a1-218f82f9675b"
      },
      "source": [
        "papersW2V.most_similar('china')"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('hubei', 0.999714195728302),\n",
              " ('late', 0.9992879629135132),\n",
              " ('rapidly', 0.9992682933807373),\n",
              " ('worldwide', 0.9992018938064575),\n",
              " ('originate', 0.999173641204834),\n",
              " ('begin', 0.9990863800048828),\n",
              " ('province', 0.9990449547767639),\n",
              " ('viral', 0.9990167617797852),\n",
              " ('recently', 0.9990154504776001),\n",
              " ('new', 0.9989810585975647)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vPVg_tm6mMwJ",
        "outputId": "90496365-3085-4406-c70e-63d9265732c5"
      },
      "source": [
        "papersW2V.most_similar('job')"
      ],
      "execution_count": 25,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('disorder', 0.9986975193023682),\n",
              " ('et', 0.9986754655838013),\n",
              " ('implement', 0.9986667633056641),\n",
              " ('search', 0.9986576437950134),\n",
              " ('numb', 0.9986507892608643),\n",
              " ('adopt', 0.9986448287963867),\n",
              " ('use', 0.9986392259597778),\n",
              " ('ct', 0.998621940612793),\n",
              " ('role', 0.9986189603805542),\n",
              " ('investigate', 0.9986182451248169)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 25
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "z2VOKXIimP5x",
        "outputId": "c2febd27-ee1d-4129-bca0-2e9b715a54fc"
      },
      "source": [
        "papersW2V.most_similar('leader')"
      ],
      "execution_count": 26,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('level', 0.9997876286506653),\n",
              " ('time', 0.9997591972351074),\n",
              " ('university', 0.9997567534446716),\n",
              " ('government', 0.9997553825378418),\n",
              " ('work', 0.9997516870498657),\n",
              " ('group', 0.9997513294219971),\n",
              " ('learn', 0.9997508525848389),\n",
              " ('community', 0.9997485876083374),\n",
              " ('lockdown', 0.9997460246086121),\n",
              " ('policy', 0.9997444152832031)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 26
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ovvJw4ComSF5",
        "outputId": "5bf0a72e-545b-429e-d8e7-8c88d241cf27"
      },
      "source": [
        "papersW2V.most_similar('economy')"
      ],
      "execution_count": 27,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('impact', 0.9998507499694824),\n",
              " ('crisis', 0.9998484253883362),\n",
              " ('society', 0.9998253583908081),\n",
              " ('work', 0.9998248815536499),\n",
              " ('hospital', 0.999822199344635),\n",
              " ('lockdown', 0.9998174905776978),\n",
              " ('practice', 0.9998120665550232),\n",
              " ('university', 0.9998114109039307),\n",
              " ('time', 0.9998106956481934),\n",
              " ('service', 0.9998049736022949)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 27
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5I7sP9HLmWVs",
        "outputId": "d9dfb9c7-b135-4569-8ec4-09dd48d5117c"
      },
      "source": [
        "papersW2V.most_similar('people')"
      ],
      "execution_count": 28,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('include', 0.9999063611030579),\n",
              " ('increase', 0.9998904466629028),\n",
              " ('large', 0.9998719692230225),\n",
              " ('early', 0.9998691082000732),\n",
              " ('understand', 0.9998657703399658),\n",
              " ('adult', 0.9998646974563599),\n",
              " ('potential', 0.9998599290847778),\n",
              " ('effective', 0.9998580813407898),\n",
              " ('strategy', 0.9998574256896973),\n",
              " ('model', 0.9998565912246704)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 28
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "E-ENakMUlfm9",
        "outputId": "ae5a3cf5-f30f-440b-83fc-2b9e278c1e24"
      },
      "source": [
        "numWords = 50\n",
        "targetWords = papersW2V.wv.index2word[:numWords]\n",
        "\n",
        "wordsSubMatrix = []\n",
        "for word in targetWords:\n",
        "    wordsSubMatrix.append(papersW2V[word])\n",
        "wordsSubMatrix = np.array(wordsSubMatrix)\n",
        "wordsSubMatrix"
      ],
      "execution_count": 18,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "array([[-0.14979531,  0.2669175 ,  0.12955476, ...,  0.11501095,\n",
              "         0.14630908,  0.29235613],\n",
              "       [-0.24249135,  0.20138872,  0.18974334, ..., -0.08300783,\n",
              "         0.2026029 ,  0.1714743 ],\n",
              "       [-0.1907625 ,  0.21985237,  0.1683869 , ...,  0.01276754,\n",
              "         0.17862743,  0.22623345],\n",
              "       ...,\n",
              "       [-0.12222155,  0.28314623,  0.11710867, ...,  0.15160415,\n",
              "         0.13312975,  0.31924036],\n",
              "       [-0.15002699,  0.30869752,  0.12305681, ...,  0.17250948,\n",
              "         0.13895474,  0.34231085],\n",
              "       [-0.13575284,  0.27041823,  0.11949792, ...,  0.145287  ,\n",
              "         0.13111664,  0.3129912 ]], dtype=float32)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 18
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "LlOltDMdlocr"
      },
      "source": [
        "pcaWords = sklearn.decomposition.PCA(n_components = 50).fit(wordsSubMatrix)\n",
        "reducedPCA_data = pcaWords.transform(wordsSubMatrix)\n",
        "# TSNE\n",
        "tsneWords = sklearn.manifold.TSNE(n_components = 2).fit_transform(reducedPCA_data)"
      ],
      "execution_count": 19,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 846
        },
        "id": "z1Q3rxkklsWh",
        "outputId": "1499c79b-c535-43c7-b5e9-e75407f74423"
      },
      "source": [
        "fig = plt.figure(figsize = (15,15))\n",
        "ax = fig.add_subplot(111)\n",
        "ax.set_frame_on(False)\n",
        "plt.scatter(tsneWords[:, 0], tsneWords[:, 1], alpha = 0)#Making the points invisible \n",
        "for i, word in enumerate(targetWords):\n",
        "    ax.annotate(word, (tsneWords[:, 0][i],tsneWords[:, 1][i]), size =  20 * (numWords - i) / numWords)\n",
        "plt.xticks(())\n",
        "plt.yticks(())\n",
        "plt.show()"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x1080 with 1 Axes>"
            ]
          },
          "metadata": {
            "tags": []
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Vg5qeQmG2BKj"
      },
      "source": [
        "# Linguistic Change"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "vB7sIE0K2FHb"
      },
      "source": [
        "import copy"
      ],
      "execution_count": 30,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "yEkwS29w2In8"
      },
      "source": [
        "def calc_syn0norm(model):\n",
        "    \"\"\"since syn0norm is now depricated\"\"\"\n",
        "    return (model.wv.syn0 / np.sqrt((model.wv.syn0 ** 2).sum(-1))[..., np.newaxis]).astype(np.float32)\n",
        "\n",
        "def smart_procrustes_align_gensim(base_embed, other_embed, words=None):\n",
        "    \"\"\"Procrustes align two gensim word2vec models (to allow for comparison between same word across models).\n",
        "    Code ported from HistWords <https://github.com/williamleif/histwords> by William Hamilton <wleif@stanford.edu>.\n",
        "    (With help from William. Thank you!)\n",
        "    First, intersect the vocabularies (see `intersection_align_gensim` documentation).\n",
        "    Then do the alignment on the other_embed model.\n",
        "    Replace the other_embed model's syn0 and syn0norm numpy matrices with the aligned version.\n",
        "    Return other_embed.\n",
        "    If `words` is set, intersect the two models' vocabulary with the vocabulary in words (see `intersection_align_gensim` documentation).\n",
        "    \"\"\"\n",
        "    base_embed = copy.copy(base_embed)\n",
        "    other_embed = copy.copy(other_embed)\n",
        "    # make sure vocabulary and indices are aligned\n",
        "    in_base_embed, in_other_embed = intersection_align_gensim(base_embed, other_embed, words=words)\n",
        "\n",
        "    # get the embedding matrices\n",
        "    base_vecs = calc_syn0norm(in_base_embed)\n",
        "    other_vecs = calc_syn0norm(in_other_embed)\n",
        "\n",
        "    # just a matrix dot product with numpy\n",
        "    m = other_vecs.T.dot(base_vecs) \n",
        "    # SVD method from numpy\n",
        "    u, _, v = np.linalg.svd(m)\n",
        "    # another matrix operation\n",
        "    ortho = u.dot(v) \n",
        "    # Replace original array with modified one\n",
        "    # i.e. multiplying the embedding matrix (syn0norm)by \"ortho\"\n",
        "    other_embed.wv.syn0norm = other_embed.wv.syn0 = (calc_syn0norm(other_embed)).dot(ortho)\n",
        "    return other_embed\n",
        "    \n",
        "def intersection_align_gensim(m1,m2, words=None):\n",
        "    \"\"\"\n",
        "    Intersect two gensim word2vec models, m1 and m2.\n",
        "    Only the shared vocabulary between them is kept.\n",
        "    If 'words' is set (as list or set), then the vocabulary is intersected with this list as well.\n",
        "    Indices are re-organized from 0..N in order of descending frequency (=sum of counts from both m1 and m2).\n",
        "    These indices correspond to the new syn0 and syn0norm objects in both gensim models:\n",
        "        -- so that Row 0 of m1.syn0 will be for the same word as Row 0 of m2.syn0\n",
        "        -- you can find the index of any word on the .index2word list: model.index2word.index(word) => 2\n",
        "    The .vocab dictionary is also updated for each model, preserving the count but updating the index.\n",
        "    \"\"\"\n",
        "\n",
        "    # Get the vocab for each model\n",
        "    vocab_m1 = set(m1.wv.vocab.keys())\n",
        "    vocab_m2 = set(m2.wv.vocab.keys())\n",
        "\n",
        "    # Find the common vocabulary\n",
        "    common_vocab = vocab_m1&vocab_m2\n",
        "    if words: common_vocab&=set(words)\n",
        "\n",
        "    # If no alignment necessary because vocab is identical...\n",
        "    if not vocab_m1-common_vocab and not vocab_m2-common_vocab:\n",
        "        return (m1,m2)\n",
        "\n",
        "    # Otherwise sort by frequency (summed for both)\n",
        "    common_vocab = list(common_vocab)\n",
        "    common_vocab.sort(key=lambda w: m1.wv.vocab[w].count + m2.wv.vocab[w].count,reverse=True)\n",
        "\n",
        "    # Then for each model...\n",
        "    for m in [m1,m2]:\n",
        "        # Replace old syn0norm array with new one (with common vocab)\n",
        "        indices = [m.wv.vocab[w].index for w in common_vocab]\n",
        "        old_arr = calc_syn0norm(m)\n",
        "        new_arr = np.array([old_arr[index] for index in indices])\n",
        "        m.wv.syn0norm = m.wv.syn0 = new_arr\n",
        "\n",
        "        # Replace old vocab dictionary with new one (with common vocab)\n",
        "        # and old index2word with new one\n",
        "        m.index2word = common_vocab\n",
        "        old_vocab = m.wv.vocab\n",
        "        new_vocab = {}\n",
        "        for new_index,word in enumerate(common_vocab):\n",
        "            old_vocab_obj=old_vocab[word]\n",
        "            new_vocab[word] = gensim.models.word2vec.Vocab(index=new_index, count=old_vocab_obj.count)\n",
        "        m.wv.vocab = new_vocab\n",
        "\n",
        "    return (m1,m2)"
      ],
      "execution_count": 31,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "qlGvsMH12gko"
      },
      "source": [
        "def compareModels(df, category, text_column_name='normalized_sents', sort = True, embeddings_raw={}):\n",
        "    \"\"\"If you are using time as your category sorting is important\"\"\"\n",
        "    if len(embeddings_raw) == 0:\n",
        "        embeddings_raw = rawModels(df, category, text_column_name, sort)\n",
        "    cats = sorted(set(df[category]))\n",
        "    #These are much quicker\n",
        "    embeddings_aligned = {}\n",
        "    for catOuter in cats:\n",
        "        embeddings_aligned[catOuter] = [embeddings_raw[catOuter]]\n",
        "        for catInner in cats:\n",
        "            embeddings_aligned[catOuter].append(smart_procrustes_align_gensim(embeddings_aligned[catOuter][-1], embeddings_raw[catInner]))\n",
        "    return embeddings_raw, embeddings_aligned"
      ],
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "6ph7o_Kx2kpB"
      },
      "source": [
        "def rawModels(df, category, text_column_name='normalized_sents', sort = True):\n",
        "    embeddings_raw = {}\n",
        "    cats = sorted(set(df[category]))\n",
        "    for cat in cats:\n",
        "        #This can take a while\n",
        "        print(\"Embedding {}\".format(cat), end = '\\r')\n",
        "        subsetDF = df[df[category] == cat]\n",
        "        #You might want to change the W2V parameters\n",
        "        embeddings_raw[cat] = gensim.models.word2vec.Word2Vec(subsetDF[text_column_name].sum())\n",
        "    return embeddings_raw"
      ],
      "execution_count": 35,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "xE-CWyKo2356"
      },
      "source": [
        "def getDivergenceDF(word, embeddingsDict):\n",
        "    dists = []\n",
        "    cats = sorted(set(embeddingsDict.keys()))\n",
        "    dists = {}\n",
        "    print(word)\n",
        "    for cat in cats:\n",
        "        dists[cat] = []\n",
        "        for embed in embeddingsDict[cat][1:]:\n",
        "            dists[cat].append(np.abs(1 - sklearn.metrics.pairwise.cosine_similarity(np.expand_dims(embeddingsDict[cat][0][word], axis = 0),\n",
        "                                                                             np.expand_dims(embed[word], axis = 0))[0,0]))\n",
        "    return pandas.DataFrame(dists, index = cats)"
      ],
      "execution_count": 39,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9v5LRBzl3Nb9"
      },
      "source": [
        "def findDiverence(word, embeddingsDict):\n",
        "    cats = sorted(set(embeddingsDict.keys()))\n",
        "    \n",
        "    dists = []\n",
        "    for embed in embeddingsDict[cats[0]][1:]:\n",
        "        dists.append(1 - sklearn.metrics.pairwise.cosine_similarity(np.expand_dims(embeddingsDict[cats[0]][0][word], axis = 0), np.expand_dims(embed[word], axis = 0))[0,0])\n",
        "    return sum(dists)\n",
        "\n",
        "def findMostDivergent(embeddingsDict):\n",
        "    words = []\n",
        "    for embeds in embeddingsDict.values():\n",
        "        for embed in embeds:\n",
        "            words += list(embed.wv.vocab.keys())\n",
        "    words = set(words)\n",
        "    print(\"Found {} words to compare\".format(len(words)))\n",
        "    return sorted([(w, findDiverence(w, embeddingsDict)) for w in words], key = lambda x: x[1], reverse=True)\n",
        "    "
      ],
      "execution_count": 43,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GUHkXsz52Mem",
        "outputId": "d04b518b-11c9-4dd7-bd90-8d03e86639c4"
      },
      "source": [
        "rawEmbeddings, comparedEmbeddings = compareModels(papers_df, 'year')"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            ""
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "ATOYgVeq3IDk",
        "outputId": "153beb28-006c-44b9-8cd0-421c1d7c10a9"
      },
      "source": [
        "wordDivergences = findMostDivergent(comparedEmbeddings)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Found 265 words to compare\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "87iqqi-C3A6N",
        "outputId": "e6b39220-c851-47a1-f49f-1882b31d1d30"
      },
      "source": [
        "# most divergent\n",
        "wordDivergences[:20]"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('question', 0.9772264361381531),\n",
              " ('america', 0.8928688168525696),\n",
              " ('real', 0.8140999674797058),\n",
              " ('period', 0.8040522336959839),\n",
              " ('model', 0.7849266529083252),\n",
              " ('introduction', 0.7819625735282898),\n",
              " ('well', 0.7156000733375549),\n",
              " ('era', 0.6698452234268188),\n",
              " ('threat', 0.6512431502342224),\n",
              " ('link', 0.6267843842506409),\n",
              " ('non', 0.6138435006141663),\n",
              " ('search', 0.5444912910461426),\n",
              " ('document', 0.5067097544670105),\n",
              " ('help', 0.4897412657737732),\n",
              " ('assessment', 0.4890210032463074),\n",
              " ('intensive', 0.48097503185272217),\n",
              " ('aspect', 0.4807313084602356),\n",
              " ('hospitalize', 0.47300034761428833),\n",
              " ('lung', 0.4536757469177246),\n",
              " ('therapy', 0.44734495878219604)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 45
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Kt_lquda3ViS",
        "outputId": "6df5ac11-570a-4dd9-dda5-115c62f00c77"
      },
      "source": [
        "wordDivergences[-20:]"
      ],
      "execution_count": 46,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[('disease', 0.022487878799438477),\n",
              " ('syndrome', 0.02231299877166748),\n",
              " ('practice', 0.021638453006744385),\n",
              " ('medical', 0.021146655082702637),\n",
              " ('student', 0.02090078592300415),\n",
              " ('acute', 0.020557105541229248),\n",
              " ('care', 0.02038121223449707),\n",
              " ('state', 0.019770681858062744),\n",
              " ('sars', 0.019412517547607422),\n",
              " ('cause', 0.019277334213256836),\n",
              " ('country', 0.019088447093963623),\n",
              " ('system', 0.018129825592041016),\n",
              " ('new', 0.01715749502182007),\n",
              " ('world', 0.017048776149749756),\n",
              " ('case', 0.015927553176879883),\n",
              " ('health', 0.01433563232421875),\n",
              " ('severe', 0.01371544599533081),\n",
              " ('coronavirus', 0.011662721633911133),\n",
              " ('pandemic', 0.010997951030731201),\n",
              " ('covid-19', 0.003960847854614258)]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 46
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "UomswxUp2u70",
        "outputId": "1402acec-d066-48f8-eb40-b76c6761e759"
      },
      "source": [
        "targetWord = 'china'\n",
        "\n",
        "pltDF = getDivergenceDF(targetWord, comparedEmbeddings)\n",
        "fig, ax = plt.subplots(figsize = (10, 7))\n",
        "seaborn.heatmap(pltDF, ax = ax, annot = False) #set annot True for a lot more information\n",
        "ax.set_xlabel(\"Starting year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_title(\"Yearly linguistic change for: '{}'\".format(targetWord))\n",
        "plt.show()"
      ],
      "execution_count": 40,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "china\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "iVBORw0KGgoAAAANSUhEUgAAAj4AAAG5CAYAAACObz5RAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADh0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uMy4yLjIsIGh0dHA6Ly9tYXRwbG90bGliLm9yZy+WH4yJAAAgAElEQVR4nO3deZhlVXn3/e+PRtAAgkCCTNIoRNNORJBgYiKGRCExad/EAWMCGqQzSNDXGMX4OBH1kcQh+ojGfgBBHICgxn4NgyaACSaNNDgxiHYQpVFQoEFQGbrrfv/Yu/BQ1HCqObuquvb347Wv3mftvdda59Sx6maNqSokSZL6YIv5roAkSdJcMfCRJEm9YeAjSZJ6w8BHkiT1hoGPJEnqDQMfSZLUGwY+WrSSHJxk3SY+e12S32rP/zbJSaOt3aRl/nqSazbx2XOTHDnCurw5yUdHlV9Xkjw2yVeS3JHk2HmsRyXZZ4prL07yubmuk6TJGfhoziX5aJIPT0h7RpJbkuw6X/WaSlW9vapeNgfl/GdVPXam+yYLSqrqsKo6rbvaLVivAS6squ2q6n2jzjzJRUkOfjB5VNXHqupZI6rPdUmWjiIvqa8MfDQfXgEcluS3AZI8FPi/wF9X1fdHUUCSLUeRjxa8vYArN+VBvyNSPxn4aM5V1S3AXwErk2wDvAn4n6o6NclBSf4ryW1Jvjr4X9tJXprk6rZb49okfzZw7eAk65K8NsmNwMQWpb9J8skJae9L8t6Z6jvYwpJkadutcWSS7ya5OcnrB+59WJLTkqxv6/qawe62iV0iSU5N8tbB9zBw7bVJbmjf7zVJDklyKPC3wAuT3Jnkq+29FyV52cCzRw98VlclecoU7+3xST6f5NYkNyX524HLWyX5SJvHlUkOGHjuuCT/M5D//zNw7SVJLk7yzvZz+HaSwwau753kP9pn/y3JiYMtWNN9BybU/QLgmcD728/iF5Ns39b5h0m+k+R/JdlioF5fTPKeJLcAb54s36kkWZKm23P8fV+WZM+BW34rybfaep+YJIOfx0A+leTPp7j3MUkuSNP6eXOSjyXZYTb1lDSDqvLwmJcD+CSwCrgF2BPYvT3/HZqg/Lfb1z/f3v+7wGOAAM8AfgI8pb12MLABOAHYGnhYm7auvb4r8GNgh/b1lsAPgP2nqNt1wG+1528GPtqeLwWKpoXqYcCTgbuBX2qvvwP4AvAIYA/ga+N1aK8XsM/A61OBtw68h/H6Pha4HthtoNzHTKzPQD4XAS9rz58P3AA8tf2s9gH2muQ9bgd8H/hr4KHt618ZKOOu9mexBPjfwOqBZ58P7Nb+nF7Yfra7ttdeAtwLHN0++xfA94C01/8beCewFfB04EcDn++034FJ3sN977t9/RHgM+17WQp8EzhqoF4baILuLduf3x8BXxvy+/o3wNfbn03an/1OAz/XzwI7AI8CfggcOlDuxRO+A1Pdu0/7nrcGfh74D+Af5/v/qx4ei+mwxUfz6S+B3wSOr6rrgT8Gzqmqc6pqrKo+D6yh+SNIVf1rVf1PNb4AfA749YH8xoA3VdXdVfXTwYKq6UL7D5o/2ACHAjdX1WWbWPe3VNVPq+qrwFdp/ggCvAB4e1Wtr6p1wKaOO9lI88dvWZKHVNV1VfU/Qz77MuDvq+rS9rNaW1XfmeS+5wA3VtW7ququqrqjqi4ZuH5x+7PYCJzOz94jVfXPVfW99ud0JvAt4MCBZ79TVf+3ffY0msBzlySPognI3lhV91TVxTTB77hpvwPTSbIEOBx4XftergPeBfzJwG3fq6r/U1Ub2p/fx6vqSTPl3XoZ8L+q6pr2c/1qNa2X495RVbdV1XeBC4H9pslr0nvbn9Xn2+/wD4F30wT5kkbEwEfzpqpuAm7mZ2M09gKe3zb/35bkNpoWgV0BkhyWZHXbLXMbzR/DnQey/GFV3TVNkafR/GGl/ff0B1H9GwfOfwJs257vRtNSM27wfGhVtRZ4JU3Lyw+SnJFktyEf3xMYJkia6b6J7/GhacfFJDkizWyq8Z/TE7j/z+K+Z6vqJ+3ptjSfz60DaXD/z2ja78AMdgYeAgwGed+haUWarKzZmu3nte1UN051b5Jd2p/1DUl+BHyU+3+ukh4kAx8tJNcDp1fVDgPHNlX1jiRb03SNvRPYpap2AM6h6XIYVzPk/y/Ak5I8gaa142MdvIfv03RxjdtzwvWfAD838PqRU2XUtkY8nSYYKJpuPJj5fV5P0yU4k+uBRw9x3/0k2Yumq+8Ymq6eHYAruP/PYirfB3ZMMvgZDH5GU34Hhsj7Zpoutr0G0h5F0+03bqbPbjrDfq4Pxttp6vjEqno4TYA+zOcqaUgGPlpIPgr8XpJntwNJH9oO+N2DZjzI1jTjITa0g2VnNUW4bQ06G/g48KW2m2HUzgJel+QRSXanCQ4GfQX4o/b9HcoU3Rhp1qf5zTbguwv4KU1XHsBNwNLxQbuTOAl4dZL909inDVYm+iywa5JXJtk6yXZJfmWI97gNzR/nH7Z1fSlNi8+M2i63NcCbk2yV5GnA7w3cMt13YKa8N9J8/m9r38tewKvaPEfhJODvkuzbfq5PSrLTiPIetx1wJ3B7+/35mxHnL/WegY8WjHacz3KaWUs/pPkv7L8BtqiqO4Bjaf6wracZlLpqiqymcxrwRB5cN9d0jgfWAd8G/o0m0Lp74PoraP7Q3wa8mKYVajJb0wyUvpmmW+QXgNe11/65/feWJJdPfLCq/hl4G02Ad0dbxo6T3HcHzUDa32vL+BbNLKlpVdVVNGNn/psmCHsi8MWZnhvwYuBpNIOW3wqcSfsZTfcdGDLvv6IZaH0tcDHNZ3DKVDenWVxw2Onw76b5/n2OZkD2yTQDpEfpLcBTgNuBfwU+NeL8pd4bn2Uh9UI7uPYbwCOr6kdzUN5fAIdXlQNUp5DkTOAbVfWm+a6LpMXPFh/1Rts19CrgjK6CniS7Jvm1JFskeSzNVPFPd1HW5irJU9v1arZou/uWM3XLlySNlCuXqhfSLJR4E80sn0M7LGor4EPA3jTdWWcAH+iwvM3RI2m6cHai6Rb8i6r68vxWSVJf2NUlSZJ6w64uSZLUGwu2q2vLrXa3KUojdcdnXjvfVdAisuVTnzPfVdAi9JCdHz2n6zbde/O1I/tbO9d131S2+EiSpN5YsC0+kiSpY2Mb57sGc84WH0mS1Bu2+EiS1Fc1NvM9i4yBjyRJfTXWv8DHri5JktQbtvhIktRTZVeXJEnqDbu6JEmSFi9bfCRJ6iu7uiRJUm+4gKEkSdLiZYuPJEl9ZVeXJEnqDWd1SZIkLV62+EiS1FMuYChJkvrDri5JkqTFyxYfSZL6yq4uSZLUGy5gKEmStHjZ4iNJUl/Z1SVJknrDWV2SJEmLly0+kiT1lV1dkiSpN+zqkiRJWrxs8ZEkqaeq+reOj4GPJEl91cMxPnZ1SZKk3rDFR5Kkvurh4GYDH0mS+qqHXV0GPpIk9ZWblEqSJI1ekkOTXJNkbZLjJrm+dZIz2+uXJFnapu+U5MIkdyZ5/xR5r0pyxTD1MPCRJKmvamx0xzSSLAFOBA4DlgEvSrJswm1HAeurah/gPcAJbfpdwBuAV0+R9x8Adw77lg18JEnqq7Gx0R3TOxBYW1XXVtU9wBnA8gn3LAdOa8/PBg5Jkqr6cVVdTBMA3U+SbYFXAW8d9i0b+EiSpK7tDlw/8HpdmzbpPVW1Abgd2GmGfP8OeBfwk2ErYuAjSVJfjbCrK8mKJGsGjhVdVj3JfsBjqurTs3nOWV2SJPXVCNfxqaqVwMopLt8A7Dnweo82bbJ71iXZEtgeuGWaIp8GHJDkOpp45heSXFRVB09XT1t8JElS1y4F9k2yd5KtgMOBVRPuWQUc2Z4/D7igqmqqDKvqg1W1W1UtBZ4OfHOmoAds8ZEkqb/maOXmqtqQ5BjgfGAJcEpVXZnkeGBNVa0CTgZOT7IWuJUmOAKgbdV5OLBVkucCz6qqqzalLgY+kiT11Fzuzl5V5wDnTEh748D5XcDzp3h26Qx5Xwc8YZh62NUlSZJ6wxYfSZL6yk1KJUlSb/Rwk1K7uiRJUm/Y4iNJUl/Z1SVJknrDri5JkqTFyxYfSZL6yq4uSZLUG3Z1SZIkLV62+EiS1Fd2dUmSpN7oYeBjV5ckSeoNW3wkSeqrHg5uNvCRJKmv7OqSJElavGzxkSSpr+zqkiRJvWFXlyRJ0uLVaYtPkl2A3duXN1TVTV2WJ0mSZsGurtFIsh/wT8D2wA1t8h5JbgP+sqou76JcSZI0Cz3s6uqqxedU4M+q6pLBxCQHAR8GnjzZQ0lWACsAsmR7tthim46qJ0mS+qirwGebiUEPQFWtTjJlNFNVK4GVAFtutXt1VDdJkgS2+IzQuUn+FfgIcH2btidwBHBeR2VKkqTZqP61MXQS+FTVsUkOA5YzMLgZOLGqzumiTEmSpJl0Nqurqs4Fzu0qf0mS9CD1sKtrztfxaQcwS5Kk+TY2NrpjMzEfCxhmHsqUJEma+y0rqupDc12mJEmahAsYjk6SRwN/QDObayPwTeDjVfWjrsqUJEmzsBl1UY1KJ11dSY6lWbn5ocBTga1pAqDVSQ7uokxJkqSZdNXiczSwX1VtTPJu4JyqOjjJh4DPAL/cUbmSJGlYruMz8rw30rT2bAtQVd9N8pAOy5QkScPqYVdXV4HPScClSS4Bfh04ASDJzwO3dlSmJEnStLpaufm9Sf4N+CXgXVX1jTb9h8BvdFGmJEmaJVt8RqeqrgSunJieZNuqurOrciVJ0pB6OJ19PhYwvGoeypQkSeqmxSfJq6a6RDvQWZIkza8ac1bXqLwd+AdgwyTX5qOVSZIkTeQYn5G5HPiXqrps4oUkL+uoTEmSpGl1Ffi8lKmnrR/QUZmSJGk2HNw8GlV1TTt1fbJrN3VRpiRJmqWxGt0xgySHJrkmydokx01yfeskZ7bXL0mytE3fKcmFSe5M8v6B+38uyb8m+UaSK5O8Y5i33NVeXdsneUdbmVuT3JLk6jZthy7KlCRJC1OSJcCJwGHAMuBFSZZNuO0oYH1V7QO8h3bxY+Au4A3AqyfJ+p1V9TiarbB+LclhM9Wlq4HGZwHrgYOraseq2gl4Zpt2VkdlSpKk2RgbG90xvQOBtVV1bVXdA5wBLJ9wz3LgtPb8bOCQJKmqH1fVxTQB0H2q6idVdWF7fg/N+OI9ZqpIV4HP0qo6oapuHKjgjVV1ArBXR2VKkqTZGGHgk2RFkjUDx4qBknYHrh94va5NY7J7qmoDcDuw0zBvo+1N+j3g32e6t6vBzd9J8hrgtPExPUl2AV7C/d+4JEmaLyPcnb2qVgIrR5bhkJJsCXwCeF9VXTvT/V21+LyQJkr7QpL1SW4FLgJ2BF7QUZmSJGlhugHYc+D1Hm3apPe0wcz2wC1D5L0S+FZV/eMwFelqk9L1ST4MfB5YPbg3V5JDgfO6KFeSJM3C3C1geCmwb5K9aQKcw4E/mnDPKuBI4L+B5wEXVE3fJJXkrTQB0tBrBHY1q+tY4DPAMcAVSQYHML29izIlSdIszdF09nbMzjHA+cDVwFlVdWWS45P8fnvbycBOSdYCrwLum/Ke5Drg3cBLkqxLsizJHsDraWaJXZ7kK8MsktzVGJ+jgf2r6s52Hv7ZSZZW1Xtp9uuSJEk9UlXnAOdMSHvjwPldwPOneHbpFNnOOqboKvDZYrx7q6quS3IwTfCzFwY+kiQtDK7cPDI3Jdlv/EUbBD0H2Bl4YkdlSpKk2ZjDlZsXiq4CnyOAGwcTqmpDVR0B/EZHZUqSJE2rq1ld66a59sUuypQkSbNTczera8HoaoyPJEla6DajLqpR6aqrS5IkacGxxUeSpL7q4awuAx9JkvrKri5JkqTFyxYfSZL6ylldkiSpN+zqkiRJWrxs8ZEkqa+c1SVJknrDri5JkqTFyxYfSZJ6yr26JElSf9jVJUmStHjZ4iNJUl/1sMXHwEeSpL7q4XR2u7okSVJv2OIjSVJf2dUlSZL6onoY+NjVJUmSesMWH0mS+qqHLT4GPpIk9VUPV262q0uSJPWGLT6SJPWVXV2SJKk3ehj42NUlSZJ6wxYfSZJ6qqp/LT4GPpIk9ZVdXZIkSYuXLT6SJPVVD1t8Fmzgc8dnXjvfVdAis93yE+a7ClpElmzxD/NdBS1Cd991/ZyW515dkiRJi9iCbfGRJEkd62GLj4GPJEl91b+tuuzqkiRJ/WGLjyRJPeXgZkmS1B9jNbpjBkkOTXJNkrVJjpvk+tZJzmyvX5JkaZu+U5ILk9yZ5P0Tntk/ydfbZ96XJDPVw8BHkiR1KskS4ETgMGAZ8KIkyybcdhSwvqr2Ad4DjK9BchfwBuDVk2T9QeBoYN/2OHSmuhj4SJLUV2MjPKZ3ILC2qq6tqnuAM4DlE+5ZDpzWnp8NHJIkVfXjqrqYJgC6T5JdgYdX1epqNh37CPDcmSpi4CNJUk/VWI3sSLIiyZqBY8VAUbsDg6szrmvTmOyeqtoA3A7sNE31d2/zmS7PB3BwsyRJetCqaiWwcr7rMRMDH0mS+mru1vG5Adhz4PUebdpk96xLsiWwPXDLDHnuMUOeD2BXlyRJPTXKrq4ZXArsm2TvJFsBhwOrJtyzCjiyPX8ecEE7dmfyuld9H/hRkoPa2VxHAJ+ZqSK2+EiSpE5V1YYkxwDnA0uAU6rqyiTHA2uqahVwMnB6krXArTTBEQBJrgMeDmyV5LnAs6rqKuAvgVOBhwHntse0DHwkSeqrOdyyoqrOAc6ZkPbGgfO7gOdP8ezSKdLXAE+YTT0MfCRJ6qnq4V5dBj6SJPVVDwMfBzdLkqTesMVHkqSesqtLkiT1Rw8DH7u6JElSb9jiI0lST9nVJUmSeqOPgY9dXZIkqTds8ZEkqaf62OJj4CNJUl9V5rsGc86uLkmS1Bu2+EiS1FN2dUmSpN6oMbu6JEmSFi1bfCRJ6im7uiRJUm+Us7okSZIWL1t8JEnqKbu6JElSbzirS5IkaRGzxUeSpJ6qmu8azD0DH0mSesquLkmSpEXMFh9Jknqqjy0+Bj6SJPVUH8f4TNvVlWRJknfOVWUkSZK6NG2LT1VtTPL0uaqMJEmaO3Z1Te7LSVYB/wz8eDyxqj7VWa0kSVLn+rhX1zCBz0OBW4DfHEgrwMBHkiRtVmYMfKrqpXNREUmSNLfcq2sSSR4KHAU8nqb1B4Cq+tMO6yVJkjo21sOurmEWMDwdeCTwbOALwB7AHV1WSpIkqQvDBD77VNUbgB9X1WnA7wK/0m21JElS16oysmNzMczg5nvbf29L8gTgRuAXuquSJEmaC05nn9zKJI8A3gCsArYF3thprSRJkjowzKyuk9rTLwCP7rY6kiRprrhlxSSS7JLk5CTntq+XJTmq+6pJkqQu1VhGdmwuhhncfCpwPrBb+/qbwCu7qpAkSVJXhgl8dq6qs4AxgKraAGzstFaSJKlzY5WRHZuLYQKfHyfZiWabCpIcBNzeaa0kSVLn5nI6e5JDk1yTZG2S4ya5vnWSM9vrlyRZOnDtdW36NUmePZD+/ya5MskVST7RLro8rWECn7+mmc31mCRfBD4C/NUQz0mSJJFkCXAicBiwDHhRkmUTbjsKWF9V+wDvAU5on10GHE6zg8ShwAeSLEmyO3AscEBVPQFY0t43rWFmdV2W5BnAY4EA11TVvTM8JkmSFrg5nNV1ILC2qq4FSHIGsBy4auCe5cCb2/OzgfcnSZt+RlXdDXw7ydo2v+/SxDEPS3Iv8HPA92aqyDCzui4DVgDfq6orDHokSVocRjnGJ8mKJGsGjhUDRe0OXD/wel2bxmT3tOOJbwd2murZqroBeCdNAPR94Paq+txM73mYrq4XtoVemuSMJM9uIzBJkiQAqmplVR0wcKzssrx2ceXlwN40M8+3SfLHMz03Y+BTVWur6vXALwIfB04BvpPkLUl2fHDVliRJ82UOBzffAOw58HqPNm3Se5JsCWwP3DLNs78FfLuqftj2Rn0K+NWZKjJMiw9JngS8C/gH4JPA84EfARcM87wkSVp4qkZ3zOBSYN8keyfZimYQ8qoJ96wCjmzPnwdcUFXVph/ezvraG9gX+BJNF9dBSX6u7Yk6BLh6porMOLi5HeNzG3AycFw7uAjgkiS/NsTzOwJU1a0z3StJkhafqtqQ5BiaBZGXAKdU1ZVJjgfWVNUqmjjj9Hbw8q20M7Ta+86iGQi9AXh5VW2kiUPOBi5v078MzNi9lpohTEvy6PFR2MNK8ijg72mir9toZoM9nKaF6Liqum6mPH567vt6uIOIurTd8hPmuwpaRJZsMVSDuTQrd991/ZyOoV2zx3NH9rf2gHX/slmM/x1mjM+sgp7WmcCngUdW1b7tnPxdgX8BzpjqocER4Sef+1+bUKwkSRrWXC5guFB09Z8sO1fVmW1TFABVtbGqzqCZmjapwRHhRx024/gkSZKkWZlxjM8muizJB4DT+Nnc+z1pBi19uaMyJUnSLGxOe2yNypSBT5I/mO7BqvrUNJePoFl6+i38bIGidcD/RzN4SZIkzbM+DqadrsXn96a5VjTz5Se/WHUP8MH2kCRJC5AtPgOq6qVdFJjkOVX12S7yliRJms5QY3yS/C7Nrqj3bfdeVcdvYplPBQx8JEmaZ5vTbKxRGWYBw3+i2fH0mcBJNKspfmlTC6yqN23qs5IkaXTG5rsC82CYFp9fraonJflaVb0lybuAc2d6KMnjaDYPGx/cfAOwqqpmXE5akiSpC8Os4/PT9t+fJNkNuJdmMcIpJXktzUKFoWkd+lJ7/okkx216dSVJ0qgUGdmxuRimxeezSXag2aD0cpoZXSfN8MxRwOPb3VLvk+TdwJXAOzahrpIkaYTGejiffcbAp6r+rj39ZJLPAg+tqttneGwM2A34zoT0Xelnl6IkSVoAhp3V9avA0vH7k1BVH5nmkVcC/57kW/xs5eZHAfsAx2xybSVJ0siMbUZdVKMyzKyu04HHAF8BxvfeKmDKwKeqzkvyi8CB3H9w86WD+3dJkqT5szmNzRmVYVp8DgCWVdWsegKragxYvUm1kiRJ6sAws7quAB45m0yTPCnJ6iTXJ1mZ5BED1zZ5DSBJkjQ6YyM8NhfDtPjsDFzVBix3jydW1e9P88wHgDfTtPi8DLg4ye9X1f8AD9n06kqSpFGxq2tyb96EfLerqvPa83cmuQw4L8mf0M/NYCVJ0gIwzHT2L2xKxkm2H5/2XlUXJvlD4JPAjpuSnyRJGq3NqYtqVKYc45Pk4vbfO5L8aOC4I8mPZsj3BOCXBhOq6mvAIcCnHmylJUnSg+cYn/t7MUBVbTfbTKvq41Okfxc4erb5SZIkjcJ0s7o+PX6S5JOzyTTJ9knekeQbSW5NckuSq9u0HTa5tpIkaWT6uFfXdIHP4Lt49CzzPQtYDxxcVTtW1U7AM9u0s2aZlyRJ6sBYRndsLqYLfGqK82EsraoTqurG+zKourGqTgD2mmVekiRJIzHdGJ8nt4OYAzxsYEBzgKqqh0/z7HeSvAY4rapuAkiyC/ASfrZ3lyRJmkfu1TWgqpY8iHxfCBwHfKENeAq4CVgFvOBB5CtJkkakjwvrDbU7+2xV1fokHwY+D6yuqjvHryU5FDhvyoclSZI6MsxeXbOW5FjgM8AxwBVJlg9cfnsXZUqSpNlxHZ/RORrYv6ruTLIUODvJ0qp6L/SwQ1GSpAVoLP37k9xV4LPFePdWVV2X5GCa4GcvDHwkSdI86aSrC7gpyX7jL9og6Dk0O70/saMyJUnSLNQIj81FVy0+RwAbBhOqagNwRJIPdVSmJEmahc1pbM6odDWra900177YRZmSJEkz6arFR5IkLXCb01YTo2LgI0lST/Vx5eauBjdLkiQtOLb4SJLUU5vTbKxRMfCRJKmn+jjGx64uSZLUG7b4SJLUU31cx8cWH0mSemouV25OcmiSa5KsTXLcJNe3TnJme/2Sdq/P8Wuva9OvSfLsgfQdkpyd5BtJrk7ytJnqYeAjSZI6lWQJcCJwGLAMeFGSZRNuOwpYX1X7AO8BTmifXQYcDjweOBT4QJsfwHuB86rqccCTgatnqouBjyRJPTWW0R0zOBBYW1XXVtU9wBnA8gn3LAdOa8/PBg5Jkjb9jKq6u6q+DawFDkyyPfAbwMkAVXVPVd02U0UMfCRJ6qmxER5JViRZM3CsGChqd+D6gdfr2jQmu6fd3/N2YKdpnt0b+CHw4SRfTnJSkm1mes8GPpIk6UGrqpVVdcDAsbLjIrcEngJ8sKp+Gfgx8ICxQxMZ+EiS1FOjbPGZwQ3AngOv92jTJr0nyZbA9sAt0zy7DlhXVZe06WfTBELTMvCRJKmnKqM7ZnApsG+SvZNsRTNYedWEe1YBR7bnzwMuqKpq0w9vZ33tDewLfKmqbgSuT/LY9plDgKtmqojr+EiSpE5V1YYkxwDnA0uAU6rqyiTHA2uqahXNIOXTk6wFbqUJjmjvO4smqNkAvLyqNrZZ/xXwsTaYuhZ46Ux1MfCRJKmn5nIBw6o6BzhnQtobB87vAp4/xbNvA942SfpXgANmUw8DH0mSesqVmyVJkhYxW3wkSeqpYbaaWGwMfCRJ6qkhVlxedOzqkiRJvWGLjyRJPdXHwc0GPpIk9VQfAx+7uiRJUm/Y4iNJUk85q0uSJPVGH2d1GfhIktRTjvGRJElaxGzxkSSppxzjs4Bs+dTnzHcVtMgs2eIf5rsKWkQ2jvWxk0CLzVgPQx+7uiRJUm8s2BYfSZLUrT62Wxr4SJLUU/3r6LKrS5Ik9YgtPpIk9ZRdXZIkqTf6uHKzXV2SJKk3bPGRJKmn+riOj4GPJEk91b+wx64uSZLUI7b4SJLUU87qkiRJvdHHMT52dUmSpN6wxUeSpJ7qX3uPgY8kSb3VxzE+dnVJkqTesMVHkqSe6uPgZgMfSZJ6qn9hj11dkiSpR2zxkSSpp/o4uNnAR5KknqoednbZ1SVJknrDFh9JknrKri5JktQbfZzObleXJEnqDVt8JEnqqf6199jiI0lSb41RIztmkuTQJNckWZvkuEmub53kzPb6JUmWDlx7XZt+TZJnT3huSZIvJ/nsMO/ZwEeSJHUqyRLgROAwYBnwoqBhsrUAABALSURBVCTLJtx2FLC+qvYB3gOc0D67DDgceDxwKPCBNr9xrwCuHrYuBj6SJPXU2AiPGRwIrK2qa6vqHuAMYPmEe5YDp7XnZwOHJEmbfkZV3V1V3wbWtvmRZA/gd4GThn3PBj6SJPVUjfB/SVYkWTNwrBgoanfg+oHX69o0JrunqjYAtwM7zfDsPwKvYRYz8x3cLEmSHrSqWgmsnKvykjwH+EFVXZbk4GGfs8VHkqSemsOurhuAPQde79GmTXpPki2B7YFbpnn214DfT3IdTdfZbyb56EwVMfCRJKmnRtnVNYNLgX2T7J1kK5rByqsm3LMKOLI9fx5wQVVVm354O+trb2Bf4EtV9bqq2qOqlrb5XVBVfzxTRezqkiRJnaqqDUmOAc4HlgCnVNWVSY4H1lTVKuBk4PQka4FbaYIZ2vvOAq4CNgAvr6qNm1qXNMHUwnPvzdcuzIpps7XtHs+Y7ypoEdk41sddjtS1DffckLks78ilfziyv7WnXffJOa37prLFR5KknhpboI0fXXKMjyRJ6g1bfCRJ6qn+tfcY+EiS1FvD7LG12NjVJUmSesMWH0mSemqI9XcWHQMfSZJ6qo+LMtjVJUmSesMWH0mSeqqPg5sNfCRJ6qk+jvGxq0uSJPWGLT6SJPVUHwc3G/hIktRTC3Wj8i7Z1SVJknrDFh9JknrKWV2SJKk3HOMjSZJ6w+nskiRJi5gtPpIk9ZRjfCRJUm84nV2SJGkRs8VHkqSeclaXJEnqDWd1SZIkLWK2+EiS1FPO6pIkSb3hrC5JkqRFrNMWnyS7ALu3L2+oqpu6LE+SJA2vj11dnbT4JNkvyWrgIuDv2+MLSVYneco0z61IsibJmpM+8okuqiZJklo1wv9tLrpq8TkV+LOqumQwMclBwIeBJ0/2UFWtBFYC3HvztZvPpyhJkjYLXQU+20wMegCqanWSbToqU5IkzcJYDwc3dxX4nJvkX4GPANe3aXsCRwDndVSmJEmahf6FPR0FPlV1bJLDgOUMDG4GTqyqc7ooU5IkaSadzeqqqnOBc7vKX5IkPTjO6poDSVbMdZmSJOmBxqiRHZuL+VjAMPNQpiRJ0txvWVFVH5rrMiVJ0gO5ZcUcSPLSuS5TkiQ9kF1dc+Mt81CmJElSN11dSb421SVgly7KlCRJs7M5bTUxKl2N8dkFeDawfkJ6gP/qqExJkjQLcznGJ8mhwHuBJcBJVfWOCde3pln4eH/gFuCFVXVde+11wFHARuDYqjo/yZ7t/bvQrMW4sqreO1M9ugp8PgtsW1VfmXghyUUdlSlJkhagJEuAE4HfBtYBlyZZVVVXDdx2FLC+qvZJcjhwAvDCJMuAw4HHA7sB/5bkF4ENwF9X1eVJtgMuS/L5CXk+QCdjfKrqqKq6eIprf9RFmZIkaXbmcHDzgcDaqrq2qu4BzqDZ3WHQcuC09vxs4JAkadPPqKq7q+rbwFrgwKr6flVdDlBVdwBX87PdIqY0H4ObJUnSAlBVIzuSrEiyZuAYXLB4d362dyc0rT4Tg5T77qmqDcDtwE7DPJtkKfDLwAM2SJ+ok8AnyZOSrE5yfZKVSR4xcO1LXZQpSZLmT1WtrKoDBo6Vc1Fukm2BTwKvrKofzXR/Vy0+HwDeDDwR+CZwcZLHtNce0lGZkiRpFuawq+sGYM+B13u0aZPek2RLYHuaQc5TPpvkITRBz8eq6lPDvOeuAp/tquq8qrqtqt4JHAOcl+Qg6OHcOUmSFqAa4f9mcCmwb5K9k2xFM1h51YR7VgFHtufPAy6oZtrZKuDwJFsn2RvYF/hSO/7nZODqqnr3sO+5sy0rkmxfVbcDVNWFSf6QJirbsasyJUnSwlNVG5IcA5xPM539lKq6MsnxwJqqWkUTxJyeZC1wK01wRHvfWcBVNDO5Xl5VG5M8HfgT4OtJxmeR/21VnTNdXdLFHP4kfwRcW1WrJ6Q/CnhDVR09Ux733nytLUMaqW33eMZ8V0GLyMaxsfmughahDffcMKcbeT9hl4NG9rf2iptWbxabkHfS4lNVH58i/bvAjEGPJEnqXh9Xbu5qVtf2Sd6R5BtJbk1yS5Kr27QduihTkiRpJl0Nbj6LZruKg6tqx6raCXhmm3ZWR2VKkqRZGKsa2bG56CrwWVpVJ1TVjeMJVXVjVZ0A7NVRmZIkaRbmcFbXgtFV4POdJK9Jct9O7El2SfJa7r/6oiRJ0pzpKvB5Ic0y019Isj7JrcBFNFPZX9BRmZIkaRb62NXV1ayu9Uk+DHweWF1Vd45fa7elP6+LciVJ0vA2py6qUelqVtexwGdoVmy+IsngDqxv76JMSZKkmXS1cvPRwP5VdWe7Y+rZSZZW1XuBzWKBI0mSFrvNqYtqVLoKfLYY796qquuSHEwT/OyFgY8kSQuCXV2jc1OS/cZftEHQc4CdaXZslyRJmnNdtfgcQbOR2H2qagNwRJIPdVSmJEmahar+7TnX1ayuddNc+2IXZUqSpNkZs6tLkiRp8eqqq0uSJC1w5awuSZLUF3Z1SZIkLWK2+EiS1FN2dUmSpN7o48rNdnVJkqTesMVHkqSe6uOWFQY+kiT1lGN8JElSbzidXZIkaRGzxUeSpJ6yq0uSJPWG09klSZIWMVt8JEnqKbu6JElSbzirS5IkaRGzxUeSpJ6yq0uSJPWGs7okSZIWMVt8JEnqKTcplSRJvWFXlyRJ0iJmi48kST3lrC5JktQbfRzjY1eXJEnqDVt8JEnqqT52ddniI0lST1XVyI6ZJDk0yTVJ1iY5bpLrWyc5s71+SZKlA9de16Zfk+TZw+Y5GQMfSZLUqSRLgBOBw4BlwIuSLJtw21HA+qraB3gPcEL77DLgcODxwKHAB5IsGTLPBzDwkSSpp2qExwwOBNZW1bVVdQ9wBrB8wj3LgdPa87OBQ5KkTT+jqu6uqm8Da9v8hsnzARbsGJ+H7PzozHcdNhdJVlTVyvmux0J3913Xz3cVNgt+nzRqfqcWrg333DCyv7VJVgArBpJWDvzcdwcGfwmvA35lQhb33VNVG5LcDuzUpq+e8Ozu7flMeT6ALT6Lw4qZb5GG5vdJo+Z3qgeqamVVHTBwLMhg18BHkiR17QZgz4HXe7Rpk96TZEtge+CWaZ4dJs8HMPCRJElduxTYN8neSbaiGay8asI9q4Aj2/PnARdUM11sFXB4O+trb2Bf4EtD5vkAC3aMj2ZlQTYnarPl90mj5neq59oxO8cA5wNLgFOq6sokxwNrqmoVcDJwepK1wK00gQztfWcBVwEbgJdX1UaAyfKcqS7p4+JFkiSpn+zqkiRJvWHgI0mSesPAZ54l2TPJhUmuSnJlkle06Tsm+XySb7X/PqJNf1yS/05yd5JXT8jrFUmuaPN55RTlJcn72uW9v5bkKd2/S82lTfhOvbj9Lnw9yX8lefJAXjMuBz/dMvNaHEb8nTolyQ+SXDFNef6eUmcMfObfBuCvq2oZcBDw8nbJ7eOAf6+qfYF/b19DM+DrWOCdg5kkeQJwNM1Klk8GnpNkn0nKO4xmRPy+NGtrfHDk70jzbbbfqW8Dz6iqJwJ/RzsQdRbLwU+6zLwWlZF8p1qn0mw7MB1/T6kzBj7zrKq+X1WXt+d3AFfTrEg5uHT3acBz23t+UFWXAvdOyOqXgEuq6idVtQH4AvAHkxS5HPhINVYDOyTZddTvS/NnE75T/1VV69v01TRrYcDwy8FPtcy8FokRfqeoqv+g+Q+46fh7Sp0x8FlA2i6CXwYuAXapqu+3l24Edpnh8SuAX0+yU5KfA36H+y/sNG6yZcN3n+Q+LQKb8J06Cji3PR/2u3K/ZeaB8WXmtQg9yO/UsPw9pc64js8CkWRb4JPAK6vqR4P/wVxVlWTadQeq6uokJwCfA34MfAXY2GGVtcDN9juV5Jk0f6SePqcV1WbD75QWA1t8FoAkD6H5ZfKxqvpUm3zTeNNu++8PZsqnqk6uqv2r6jeA9cA3J7ltk5b41uZltt+pJE8CTgKWV9UtbfKw35WplpnXIjKi79Sw/D2lzhj4zLN2LMTJwNVV9e6BS4NLdx8JfGaIvH6h/fdRNON7Pj7JbauAI9pZEwcBtw80VWsRmO13qv2+fAr4k6oaDJaHXQ5+qmXmtUiM8Ds1LH9PqTOu3DzPkjwd+E/g68BYm/y3NP3nZwGPAr4DvKCqbk3ySGAN8PD2/juBZW2z83/SjK24F3hVVf17W8afA1TVP7W/wN5PM6viJ8BLq2rNnLxZzYlN+E6dBPxhmwawoaoOaPP6HeAf+dly8G9r0+9bZj7JQ4HTacZ93AocXlXXdv9ONVdG/J36BHAwsDNwE/CmqjrZ31OaKwY+kiSpN+zqkiRJvWHgI0mSesPAR5Ik9YaBjyRJ6g0DH0mS1BsGPtIClOT17S7YX0vylSS/0qa/st2SZLb5vSTJbgOvT5piw1FJWtSczi4tMEmeBrwbOLiq7k6yM7BVVX0vyXXAAVV18yzyW0Kzc/arN5e1UJIsqSq3XJE0crb4SAvPrsDNVXU3QFXd3AY9xwK7ARcmuRAgyQeTrGlbh94ynkGS65KckORy4EXAAcDH2tajhyW5KMn4gnJ3Jnlbkq8mWZ1klzb9Me3rryd5a5I7J1Y0yfFJXjnw+m1JXtGe/02SS9tWq8G6/UuSy9o6rxhIvzPJu5J8FXjaCD9PSbqPgY+08HwO2DPJN5N8IMkzAKrqfcD3gGdW1TPbe1/froj7JOAZ7f5I426pqqdU1UdpVvt+cVXtV1U/nVDeNsDqqnoy8B/A0W36e4H3VtUTaXbHnswpwBEASbag2dbio0meBewLHAjsB+yf5DfaZ/60qvanCcaOTTK+k/s2wCVV9eSqunjoT0uSZsHAR1pgqupOYH9gBfBD4MwkL5ni9he0rTpfBh4PDI7bOXPIIu8BPtueXwYsbc+fBvxzez7Zvm9U1XXALUl+GXgW8OV2Q8pnjb8GLgceRxMIQRPsfBVYTbMR5Xj6RppNMCWpM1vOdwUkPVA7vuUi4KIkX6fZAPLUwXuS7A28GnhqVa1Pcirw0IFbfjxkcfcObCq6kdn/XjgJeAnwSJoWIIAA/7uqPjShzgcDvwU8rap+kuSigTrf5bgeSV2zxUdaYJI8Nsm+A0n78bPNHu8AtmvPH04T3Nzejss5bJpsB58b1mqajSah6cKayqdpNpN8KnB+m3Y+8KdJtgVIsnuSXwC2B9a3Qc/jgINmWSdJelBs8ZEWnm2B/5NkB2ADsJam2wtgJXBeku9V1TOTfBn4BnA98MVp8jwV+KckP2X4gcOvpBmv83rgPOD2yW6qqnvawda3jbfYVNXnkvwS8N/NRtvcCfxxm8+fJ7kauIYmuJKkOeN0dkmTatcL+mlVVZLDgRdV1fJJ7tuCZhzP86vqW3NdT0maDVt8JE1lf+D9aZpsbgP+dOIN7SKInwU+bdAjaXNgi48kSeoNBzdLkqTeMPCRJEm9YeAjSZJ6w8BHkiT1hoGPJEnqjf8fQIh3ggNd5hsAAAAASUVORK5CYII=\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "7dwVb-UK3cqt",
        "outputId": "174a7960-5b50-4e85-af89-f5b87b0eb95c"
      },
      "source": [
        "targetWord= \"america\"\n",
        "pltDF = getDivergenceDF(targetWord, comparedEmbeddings)\n",
        "fig, ax = plt.subplots(figsize = (10, 7))\n",
        "seaborn.heatmap(pltDF, ax = ax, annot = False) #set annot True for a lot more information\n",
        "ax.set_xlabel(\"Starting year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_title(\"Yearly linguistic change for: '{}'\".format(targetWord))\n",
        "plt.show()"
      ],
      "execution_count": 47,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "america\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "hUfwnTdh3gkC",
        "outputId": "25da26f6-7ad6-42b3-edca-944ab0a907b0"
      },
      "source": [
        "targetWord= \"quarantine\"\n",
        "pltDF = getDivergenceDF(targetWord, comparedEmbeddings)\n",
        "fig, ax = plt.subplots(figsize = (10, 7))\n",
        "seaborn.heatmap(pltDF, ax = ax, annot = False) #set annot True for a lot more information\n",
        "ax.set_xlabel(\"Starting year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_title(\"Yearly linguistic change for: '{}'\".format(targetWord))\n",
        "plt.show()"
      ],
      "execution_count": 48,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "quarantine\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "142-js1s3i0O",
        "outputId": "51399960-db8a-4bd4-8554-486ce873354a"
      },
      "source": [
        "targetWord= \"vaccine\"\n",
        "pltDF = getDivergenceDF(targetWord, comparedEmbeddings)\n",
        "fig, ax = plt.subplots(figsize = (10, 7))\n",
        "seaborn.heatmap(pltDF, ax = ax, annot = False) #set annot True for a lot more information\n",
        "ax.set_xlabel(\"Starting year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_title(\"Yearly linguistic change for: '{}'\".format(targetWord))\n",
        "plt.show()"
      ],
      "execution_count": 49,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "vaccine\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 475
        },
        "id": "_GLG--EG3p0B",
        "outputId": "7228f0b1-3ee8-4d7c-d3ad-a0a2e11013a5"
      },
      "source": [
        "targetWord = 'threat'\n",
        "\n",
        "pltDF = getDivergenceDF(targetWord, comparedEmbeddings)\n",
        "fig, ax = plt.subplots(figsize = (10, 7))\n",
        "seaborn.heatmap(pltDF, ax = ax, annot = False) #set annot True for a lot more information\n",
        "ax.set_xlabel(\"Starting year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_ylabel(\"Final year\")\n",
        "ax.set_title(\"Yearly linguistic change for: '{}'\".format(targetWord))\n",
        "plt.show()"
      ],
      "execution_count": 51,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "threat\n"
          ],
          "name": "stdout"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 720x504 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "hpcGvzZKhV_g"
      },
      "source": [
        "# Sentiment Analysis"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oby1KtaGygpP"
      },
      "source": [
        "# sentiment\n",
        "full_content= (\"\".join(i for i in full_df['text']))\n",
        "full_sentiment = TextBlob(full_content)"
      ],
      "execution_count": 60,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "S-1kAcNL0WVp",
        "outputId": "4b8c0894-e78a-4cc4-81a2-88be75e059ab"
      },
      "source": [
        "econ_content= (\"\".join(i for i in econ_df['text']))\n",
        "econ_sentiment = TextBlob(econ_content)\n",
        "econ_sentiment.sentiment"
      ],
      "execution_count": 63,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Sentiment(polarity=0.0728681077814532, subjectivity=0.3848585802602516)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 63
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "lE1GXYWs1EHq",
        "outputId": "c4d0c389-2e70-4979-f3c0-8592fda4cdb6"
      },
      "source": [
        "psych_content= (\"\".join(i for i in psych_df['text']))\n",
        "psych_sentiment = TextBlob(psych_content)\n",
        "psych_sentiment.sentiment"
      ],
      "execution_count": 65,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Sentiment(polarity=0.07493156434782125, subjectivity=0.3916037609531709)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 65
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UAkAfm-311oe",
        "outputId": "d700d8e8-0dfa-420a-e40d-646fe0dc8713"
      },
      "source": [
        "papers_content= (\"\".join(i for i in papers_df['text']))\n",
        "papers_sentiment = TextBlob(papers_content)\n",
        "papers_sentiment.sentiment"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Sentiment(polarity=0.10677599837734063, subjectivity=0.4166789151770683)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 67
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Zcu7NYzfz9r6",
        "outputId": "a735c27a-cddc-40b0-c166-d481eb611fc7"
      },
      "source": [
        "full_sentiment.sentiment"
      ],
      "execution_count": 62,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "Sentiment(polarity=0.07335415371042517, subjectivity=0.38644740343657424)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 62
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "Q4Is4_c_sJDg"
      },
      "source": [
        "# K-means"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "LmBlPc52sNKy"
      },
      "source": [
        "## Social Science"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "oOTvumtVv1Hn"
      },
      "source": [
        "def plotSilhouette(n_clusters, X):\n",
        "    fig, (ax1, ax2) = plt.subplots(ncols=2, figsize = (15,5))\n",
        "    \n",
        "    ax1.set_xlim([-0.1, 1])\n",
        "    ax1.set_ylim([0, len(X) + (n_clusters + 1) * 10])\n",
        "    clusterer = sklearn.cluster.KMeans(n_clusters=n_clusters, random_state=10)\n",
        "    cluster_labels = clusterer.fit_predict(X)\n",
        "    \n",
        "    silhouette_avg = sklearn.metrics.silhouette_score(X, cluster_labels)\n",
        "\n",
        "    # Compute the silhouette scores for each sample\n",
        "    sample_silhouette_values = sklearn.metrics.silhouette_samples(X, cluster_labels)\n",
        "\n",
        "    y_lower = 10\n",
        "    \n",
        "    for i in range(n_clusters):\n",
        "        ith_cluster_silhouette_values = sample_silhouette_values[cluster_labels == i]\n",
        "\n",
        "        ith_cluster_silhouette_values.sort()\n",
        "\n",
        "        size_cluster_i = ith_cluster_silhouette_values.shape[0]\n",
        "        y_upper = y_lower + size_cluster_i\n",
        "        cmap = matplotlib.cm.get_cmap(\"nipy_spectral\")\n",
        "        color = cmap(float(i) / n_clusters)\n",
        "        ax1.fill_betweenx(np.arange(y_lower, y_upper),\n",
        "                          0, ith_cluster_silhouette_values,\n",
        "                          facecolor=color, edgecolor=color, alpha=0.7)\n",
        "\n",
        "        ax1.text(-0.05, y_lower + 0.5 * size_cluster_i, str(i))\n",
        "\n",
        "        y_lower = y_upper + 10\n",
        "    \n",
        "    ax1.set_title(\"The silhouette plot for the various clusters.\")\n",
        "    ax1.set_xlabel(\"The silhouette coefficient values\")\n",
        "    ax1.set_ylabel(\"Cluster label\")\n",
        "\n",
        "    ax1.axvline(x=silhouette_avg, color=\"red\", linestyle=\"--\")\n",
        "\n",
        "    ax1.set_yticks([])  # Clear the yaxis labels / ticks\n",
        "    ax1.set_xticks([-0.1, 0, 0.2, 0.4, 0.6, 0.8, 1])\n",
        "\n",
        "    # 2nd Plot showing the actual clusters formed\n",
        "    cmap = matplotlib.cm.get_cmap(\"nipy_spectral\")\n",
        "    colors = cmap(float(i) / n_clusters)\n",
        "    ax2.scatter(reduced_data[:, 0], reduced_data[:, 1], marker='.', s=30, lw=0, alpha=0.7,\n",
        "                c=colors)\n",
        "\n",
        "    # Labeling the clusters\n",
        "    centers = clusterer.cluster_centers_\n",
        "    projected_centers = pca.transform(centers)\n",
        "    # Draw white circles at cluster centers\n",
        "    ax2.scatter(projected_centers[:, 0], projected_centers[:, 1],\n",
        "                marker='o', c=\"white\", alpha=1, s=200)\n",
        "\n",
        "    for i, c in enumerate(projected_centers):\n",
        "        ax2.scatter(c[0], c[1], marker='$%d$' % i, alpha=1, s=50)\n",
        "\n",
        "    ax2.set_title(\"The visualization of the clustered data.\")\n",
        "    ax2.set_xlabel(\"PC 1\")\n",
        "    ax2.set_ylabel(\"PC 2\")\n",
        "\n",
        "    plt.suptitle((\"Silhouette analysis for KMeans clustering on sample data \"\n",
        "                  \"with n_clusters = %d\" % n_clusters),\n",
        "                 fontsize=14, fontweight='bold')\n",
        "    plt.show()\n",
        "    print(\"For n_clusters = {}, The average silhouette_score is : {:.3f}\".format(n_clusters, silhouette_avg))"
      ],
      "execution_count": 12,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "vXu7bqabu4fY",
        "outputId": "2765a24d-4c2d-4cbe-94be-f5231abc9e59"
      },
      "source": [
        "#initialize\n",
        "full_Vectorizer = sklearn.feature_extraction.text.TfidfVectorizer(max_df=0.8, min_df=5, max_features=1000, stop_words='english', norm='l2')\n",
        "#train\n",
        "full_vects = full_Vectorizer.fit_transform(full_df['text'])\n",
        "full_vects.shape"
      ],
      "execution_count": 11,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(72, 1000)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 11
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Sp1IdU0aw3S1"
      },
      "source": [
        "PCA = sklearn.decomposition.PCA\n",
        "pca = PCA(n_components = 2).fit(papers_vects.toarray())\n",
        "reduced_data = pca.transform(full_vects.toarray())"
      ],
      "execution_count": 23,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        },
        "id": "eRqdBuHOv8f7",
        "outputId": "5bfabbec-430e-42a2-bcb6-334447d3edd2"
      },
      "source": [
        "# try 2 clusters\n",
        "X = full_vects.toarray()\n",
        "plotSilhouette(2, X)"
      ],
      "execution_count": 19,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "For n_clusters = 2, The average silhouette_score is : 0.049\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 425
        },
        "id": "DbpPXtSkw8aY",
        "outputId": "c7c9d3b6-96cb-4ba3-fab3-22268518e6a1"
      },
      "source": [
        "# try 2 clusters\n",
        "X = full_vects.toarray()\n",
        "plotSilhouette(3, X)"
      ],
      "execution_count": 22,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "For n_clusters = 3, The average silhouette_score is : 0.061\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "3wL4a8Yox-H3",
        "outputId": "78f54bb6-0185-416e-97a9-a57b7b4cfba7"
      },
      "source": [
        "full_km = sklearn.cluster.KMeans(2, init='k-means++')\n",
        "full_km.fit(full_vects)"
      ],
      "execution_count": 24,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
              "       n_clusters=2, n_init=10, n_jobs=None, precompute_distances='auto',\n",
              "       random_state=None, tol=0.0001, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 24
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "VrS8z_Jyybeu",
        "outputId": "77204bf8-7325-43ff-db0c-9458d7411862"
      },
      "source": [
        "full_terms = full_Vectorizer.get_feature_names()\n",
        "print(\"Top terms per cluster:\")\n",
        "order_centroids = full_km.cluster_centers_.argsort()[:, ::-1]\n",
        "for i in range(2):\n",
        "    print(\"Cluster %d:\" % i)\n",
        "    for ind in order_centroids[i, :10]:\n",
        "        print(' %s' % full_terms[ind])\n",
        "    print('\\n')"
      ],
      "execution_count": 29,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Top terms per cluster:\n",
            "Cluster 0:\n",
            " doi\n",
            " dx\n",
            " job\n",
            " psychological\n",
            " anxiety\n",
            " psychology\n",
            " employees\n",
            " 1037\n",
            " 01\n",
            " engagement\n",
            "\n",
            "\n",
            "Cluster 1:\n",
            " infected\n",
            " home\n",
            " policy\n",
            " employment\n",
            " county\n",
            " percent\n",
            " unemployment\n",
            " infection\n",
            " panel\n",
            " labor\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "TUrOZwMyy2L_"
      },
      "source": [
        "## Natural Science"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "qViyikD_sPuH",
        "outputId": "48c4a170-ec97-4f2b-8e50-608014b7bebf"
      },
      "source": [
        "#initialize\n",
        "papers_Vectorizer = sklearn.feature_extraction.text.TfidfVectorizer(max_df=0.8, min_df=5, max_features=1000, stop_words='english', norm='l2')\n",
        "#train\n",
        "papers_vects = papers_Vectorizer.fit_transform(papers_df['text'])\n",
        "papers_vects.shape"
      ],
      "execution_count": 32,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "(1959, 1000)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 32
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "j1wNWK4vu1RM"
      },
      "source": [
        "PCA = sklearn.decomposition.PCA\n",
        "pca = PCA(n_components = 2).fit(papers_vects.toarray())\n",
        "reduced_data = pca.transform(papers_vects.toarray())"
      ],
      "execution_count": 33,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        },
        "id": "zfINx_KPzC7l",
        "outputId": "8a9b20e3-62ad-491d-c1f2-3945023cbb0d"
      },
      "source": [
        "# try 2 clusters\n",
        "X = papers_vects.toarray()\n",
        "plotSilhouette(2, X)"
      ],
      "execution_count": 34,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "For n_clusters = 2, The average silhouette_score is : 0.008\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        },
        "id": "LLbqTob0zIUp",
        "outputId": "a8545ab4-0ff7-4354-ab37-63cfe20bd3dc"
      },
      "source": [
        "# try 2 clusters\n",
        "X = papers_vects.toarray()\n",
        "plotSilhouette(3, X)"
      ],
      "execution_count": 35,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "For n_clusters = 3, The average silhouette_score is : 0.013\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 421
        },
        "id": "ib0bMD48zWP_",
        "outputId": "5c96cc08-7596-451d-cfee-d72b1dce14ac"
      },
      "source": [
        "# try 2 clusters\n",
        "X = papers_vects.toarray()\n",
        "plotSilhouette(4, X)"
      ],
      "execution_count": 36,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "*c* argument looks like a single numeric RGB or RGBA sequence, which should be avoided as value-mapping will have precedence in case its length matches with *x* & *y*.  Please use the *color* keyword-argument or provide a 2-D array with a single row if you intend to specify the same RGB or RGBA value for all points.\n"
          ],
          "name": "stderr"
        },
        {
          "output_type": "display_data",
          "data": {
            "image/png": "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\n",
            "text/plain": [
              "<Figure size 1080x360 with 2 Axes>"
            ]
          },
          "metadata": {
            "tags": [],
            "needs_background": "light"
          }
        },
        {
          "output_type": "stream",
          "text": [
            "For n_clusters = 4, The average silhouette_score is : 0.015\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "5eEVrsrUzbQ8",
        "outputId": "541be960-5bc3-4570-ca55-2664919b9009"
      },
      "source": [
        "papers_km = sklearn.cluster.KMeans(2, init='k-means++')\n",
        "papers_km.fit(papers_vects)"
      ],
      "execution_count": 44,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
              "       n_clusters=2, n_init=10, n_jobs=None, precompute_distances='auto',\n",
              "       random_state=None, tol=0.0001, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 44
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "FuCIwYwv0Wjt",
        "outputId": "e4ecc434-10a5-42b5-8c4f-1fda2e501fd3"
      },
      "source": [
        "papers_terms = papers_Vectorizer.get_feature_names()\n",
        "print(\"Top terms per cluster:\")\n",
        "order_centroids = papers_km.cluster_centers_.argsort()[:, ::-1]\n",
        "for i in range(2):\n",
        "    print(\"Cluster %d:\" % i)\n",
        "    for ind in order_centroids[i, :10]:\n",
        "        print(' %s' % papers_terms[ind])\n",
        "    print('\\n')"
      ],
      "execution_count": 45,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Top terms per cluster:\n",
            "Cluster 0:\n",
            " pandemic\n",
            " disease\n",
            " patients\n",
            " coronavirus\n",
            " health\n",
            " 2019\n",
            " 2020\n",
            " study\n",
            " world\n",
            " clinical\n",
            "\n",
            "\n",
            "Cluster 1:\n",
            " coronavirus\n",
            " respiratory\n",
            " sars\n",
            " cov\n",
            " 2019\n",
            " acute\n",
            " severe\n",
            " syndrome\n",
            " disease\n",
            " caused\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "HKoVXT8H04MK",
        "outputId": "30eeec9b-baf7-4f10-d5ad-8a5da39adafa"
      },
      "source": [
        "papers_km = sklearn.cluster.KMeans(3, init='k-means++')\n",
        "papers_km.fit(papers_vects)"
      ],
      "execution_count": 41,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "KMeans(algorithm='auto', copy_x=True, init='k-means++', max_iter=300,\n",
              "       n_clusters=3, n_init=10, n_jobs=None, precompute_distances='auto',\n",
              "       random_state=None, tol=0.0001, verbose=0)"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 41
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "Qigw-Etm07n-",
        "outputId": "6828c260-bf68-474d-846a-b310fddcfe67"
      },
      "source": [
        "papers_terms = papers_Vectorizer.get_feature_names()\n",
        "print(\"Top terms per cluster:\")\n",
        "order_centroids = papers_km.cluster_centers_.argsort()[:, ::-1]\n",
        "for i in range(3):\n",
        "    print(\"Cluster %d:\" % i)\n",
        "    for ind in order_centroids[i, :10]:\n",
        "        print(' %s' % papers_terms[ind])\n",
        "    print('\\n')"
      ],
      "execution_count": 42,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "Top terms per cluster:\n",
            "Cluster 0:\n",
            " la\n",
            " di\n",
            " en\n",
            " da\n",
            " el\n",
            " del\n",
            " pandemia\n",
            " por\n",
            " na\n",
            " para\n",
            "\n",
            "\n",
            "Cluster 1:\n",
            " coronavirus\n",
            " respiratory\n",
            " sars\n",
            " cov\n",
            " acute\n",
            " severe\n",
            " syndrome\n",
            " 2019\n",
            " disease\n",
            " caused\n",
            "\n",
            "\n",
            "Cluster 2:\n",
            " pandemic\n",
            " disease\n",
            " patients\n",
            " coronavirus\n",
            " health\n",
            " 2019\n",
            " 2020\n",
            " study\n",
            " world\n",
            " clinical\n",
            "\n",
            "\n"
          ],
          "name": "stdout"
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "260cq9bnnVZN"
      },
      "source": [
        "# Topic Modeling"
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "PNXfRtiW8UYI"
      },
      "source": [
        "## Social Science"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "0-tf6kL5ot3i"
      },
      "source": [
        "dictionary_full = gensim.corpora.Dictionary(full_df['normalized_tokens'])"
      ],
      "execution_count": 46,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "Zs873IeXpjUf"
      },
      "source": [
        "corpus_full = [dictionary_full.doc2bow(text) for text in full_df['normalized_tokens']]"
      ],
      "execution_count": 47,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "g3la9ocypyF-"
      },
      "source": [
        "lda_full = gensim.models.ldamodel.LdaModel(corpus=corpus_full, id2word=dictionary_full, num_topics=5, alpha='auto', eta='auto')"
      ],
      "execution_count": 57,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 355
        },
        "id": "qqthNeHSqCLI",
        "outputId": "7b5b36cd-a7a0-4add-ccef-419edd7ead80"
      },
      "source": [
        "topicsDict = {}\n",
        "for topicNum in range(lda_full.num_topics):\n",
        "    topicWords = [w for w, p in lda_full.show_topic(topicNum)]\n",
        "    topicsDict['Topic_{}'.format(topicNum)] = topicWords\n",
        "\n",
        "wordRanksDF = pandas.DataFrame(topicsDict)\n",
        "wordRanksDF"
      ],
      "execution_count": 58,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Topic_0</th>\n",
              "      <th>Topic_1</th>\n",
              "      <th>Topic_2</th>\n",
              "      <th>Topic_3</th>\n",
              "      <th>Topic_4</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>rate</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>=</td>\n",
              "      <td>=</td>\n",
              "      <td>covid-19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>=</td>\n",
              "      <td>=</td>\n",
              "      <td>work</td>\n",
              "      <td>work</td>\n",
              "      <td>work</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>work</td>\n",
              "      <td>rate</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>model</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>−</td>\n",
              "      <td>policy</td>\n",
              "      <td>state</td>\n",
              "      <td>rate</td>\n",
              "      <td>=</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>t</td>\n",
              "      <td>work</td>\n",
              "      <td>model</td>\n",
              "      <td>datum</td>\n",
              "      <td>rate</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>model</td>\n",
              "      <td>datum</td>\n",
              "      <td>time</td>\n",
              "      <td>job</td>\n",
              "      <td>time</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>datum</td>\n",
              "      <td>model</td>\n",
              "      <td>datum</td>\n",
              "      <td>state</td>\n",
              "      <td>datum</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>covid-19</td>\n",
              "      <td>case</td>\n",
              "      <td>change</td>\n",
              "      <td>case</td>\n",
              "      <td>level</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>state</td>\n",
              "      <td>firm</td>\n",
              "      <td>case</td>\n",
              "      <td>time</td>\n",
              "      <td>effect</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>policy</td>\n",
              "      <td>effect</td>\n",
              "      <td>effect</td>\n",
              "      <td>figure</td>\n",
              "      <td>individual</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "    Topic_0   Topic_1   Topic_2   Topic_3     Topic_4\n",
              "0      rate  covid-19         =         =    covid-19\n",
              "1         =         =      work      work        work\n",
              "2      work      rate  covid-19  covid-19       model\n",
              "3         −    policy     state      rate           =\n",
              "4         t      work     model     datum        rate\n",
              "5     model     datum      time       job        time\n",
              "6     datum     model     datum     state       datum\n",
              "7  covid-19      case    change      case       level\n",
              "8     state      firm      case      time      effect\n",
              "9    policy    effect    effect    figure  individual"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 58
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "HXFHNj6w8X64"
      },
      "source": [
        "## Natural Science"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "1OcrUJTd8dxv"
      },
      "source": [
        "dictionary_papers = gensim.corpora.Dictionary(papers_df['normalized_tokens'])"
      ],
      "execution_count": 52,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "jMlxUO2Q8hQh"
      },
      "source": [
        "corpus_papers = [dictionary_papers.doc2bow(text) for text in papers_df['normalized_tokens']]"
      ],
      "execution_count": 53,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "zqiah3GH8odB"
      },
      "source": [
        "lda_papers = gensim.models.ldamodel.LdaModel(corpus=corpus_papers, id2word=dictionary_papers, num_topics=5, alpha='auto', eta='auto')"
      ],
      "execution_count": 56,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 355
        },
        "id": "FauWbeaj8vE4",
        "outputId": "b675541a-7b64-4a72-dfc0-507c47efd280"
      },
      "source": [
        "topicsDict = {}\n",
        "for topicNum in range(lda_papers.num_topics):\n",
        "    topicWords = [w for w, p in lda_papers.show_topic(topicNum)]\n",
        "    topicsDict['Topic_{}'.format(topicNum)] = topicWords\n",
        "\n",
        "wordRanksDF = pandas.DataFrame(topicsDict)\n",
        "wordRanksDF"
      ],
      "execution_count": 59,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>Topic_0</th>\n",
              "      <th>Topic_1</th>\n",
              "      <th>Topic_2</th>\n",
              "      <th>Topic_3</th>\n",
              "      <th>Topic_4</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>covid-19</td>\n",
              "      <td>de</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>pandemic</td>\n",
              "      <td>la</td>\n",
              "      <td>disease</td>\n",
              "      <td>coronavirus</td>\n",
              "      <td>coronavirus</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>coronavirus</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>coronavirus</td>\n",
              "      <td>disease</td>\n",
              "      <td>disease</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>case</td>\n",
              "      <td>coronavirus</td>\n",
              "      <td>pandemic</td>\n",
              "      <td>pandemic</td>\n",
              "      <td>sars</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>disease</td>\n",
              "      <td>covid?\\19</td>\n",
              "      <td>patient</td>\n",
              "      <td>health</td>\n",
              "      <td>patient</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>health</td>\n",
              "      <td>sars</td>\n",
              "      <td>severe</td>\n",
              "      <td>patient</td>\n",
              "      <td>cause</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>outbreak</td>\n",
              "      <td>outbreak</td>\n",
              "      <td>health</td>\n",
              "      <td>infection</td>\n",
              "      <td>respiratory</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>covid?\\19</td>\n",
              "      <td>en</td>\n",
              "      <td>sars</td>\n",
              "      <td>china</td>\n",
              "      <td>cov-2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>global</td>\n",
              "      <td>le</td>\n",
              "      <td>respiratory</td>\n",
              "      <td>outbreak</td>\n",
              "      <td>china</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>cov-2</td>\n",
              "      <td>cov-2</td>\n",
              "      <td>acute</td>\n",
              "      <td>severe</td>\n",
              "      <td>pandemic</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       Topic_0      Topic_1      Topic_2      Topic_3      Topic_4\n",
              "0     covid-19           de     covid-19     covid-19     covid-19\n",
              "1     pandemic           la      disease  coronavirus  coronavirus\n",
              "2  coronavirus     covid-19  coronavirus      disease      disease\n",
              "3         case  coronavirus     pandemic     pandemic         sars\n",
              "4      disease    covid?\\19      patient       health      patient\n",
              "5       health         sars       severe      patient        cause\n",
              "6     outbreak     outbreak       health    infection  respiratory\n",
              "7    covid?\\19           en         sars        china        cov-2\n",
              "8       global           le  respiratory     outbreak        china\n",
              "9        cov-2        cov-2        acute       severe     pandemic"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 59
        }
      ]
    },
    {
      "cell_type": "markdown",
      "metadata": {
        "id": "i6vOfP4TGMAo"
      },
      "source": [
        "# Dynamic Topic Modeling"
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "lMMeYN2MNAhd"
      },
      "source": [
        "from gensim.corpora import Dictionary, bleicorpus\n",
        "from gensim.matutils import hellinger"
      ],
      "execution_count": 65,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "BItGVDJUGInv"
      },
      "source": [
        "time_slice = [640, 836, 483]"
      ],
      "execution_count": 61,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "T5Lmn2NtGTLd",
        "outputId": "29647242-4f98-4b7e-b8e1-8b4e25e67a21"
      },
      "source": [
        "dictionary_seq = gensim.corpora.Dictionary(papers_df['normalized_tokens'])\n",
        "#dictionary = gensim.corpora.Dictionary(final_corpus)\n",
        "corpus_seq = [dictionary_seq.doc2bow(text) for text in papers_df['normalized_tokens']]\n",
        "ldaseq = ldaseqmodel.LdaSeqModel(corpus=corpus_seq, id2word=dictionary_seq, time_slice=time_slice, num_topics=2)"
      ],
      "execution_count": 67,
      "outputs": [
        {
          "output_type": "stream",
          "text": [
            "/usr/local/lib/python3.7/dist-packages/gensim/models/ldaseqmodel.py:290: RuntimeWarning: divide by zero encountered in double_scalars\n",
            "  convergence = np.fabs((bound - old_bound) / old_bound)\n"
          ],
          "name": "stderr"
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "cymKLmQVHEFj",
        "outputId": "c655f963-4ed3-48d1-bdf0-a96f0ac73c6d"
      },
      "source": [
        "ldaseq.print_topics(time=0)"
      ],
      "execution_count": 68,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('covid-19', 0.046723443618186804),\n",
              "  ('pandemic', 0.01666185694081015),\n",
              "  ('de', 0.013147735282885771),\n",
              "  ('la', 0.00805385891530807),\n",
              "  ('impact', 0.0064823056876690205),\n",
              "  ('crisis', 0.004873021391956587),\n",
              "  ('e', 0.004615595950721479),\n",
              "  ('global', 0.004246331244818657),\n",
              "  ('health', 0.004170162839912912),\n",
              "  ('social', 0.003980834727989875),\n",
              "  ('en', 0.003974607484036784),\n",
              "  ('di', 0.0038445581715665614),\n",
              "  ('time', 0.0034011534113324473),\n",
              "  ('article', 0.0033993178606789775),\n",
              "  ('economic', 0.0033862981987582675),\n",
              "  ('covid', 0.0033821781033152687),\n",
              "  ('world', 0.0032863856979313067),\n",
              "  ('medical', 0.003206116218973925),\n",
              "  ('new', 0.0030615651854937033),\n",
              "  ('learn', 0.003056385709554595)],\n",
              " [('covid-19', 0.06346346053327592),\n",
              "  ('coronavirus', 0.02673126389515641),\n",
              "  ('disease', 0.02401511538115626),\n",
              "  ('patient', 0.01330413138984548),\n",
              "  ('pandemic', 0.012128038912016657),\n",
              "  ('sars', 0.010308268450830911),\n",
              "  ('health', 0.00986025848576333),\n",
              "  ('severe', 0.009223478843966168),\n",
              "  ('respiratory', 0.008958171431812089),\n",
              "  ('cov-2', 0.008903325532338982),\n",
              "  ('case', 0.00882721770759777),\n",
              "  ('acute', 0.008104757633943211),\n",
              "  ('china', 0.007967753332214873),\n",
              "  ('cause', 0.007711476302335946),\n",
              "  ('infection', 0.007567542583296424),\n",
              "  ('outbreak', 0.007113328763442282),\n",
              "  ('covid?\\\\19', 0.006904749978605111),\n",
              "  ('syndrome', 0.006592741782246652),\n",
              "  ('study', 0.006369879843431602),\n",
              "  ('clinical', 0.006069589891108472)]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 68
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "UhDvxHrsSoAx",
        "outputId": "43721454-dec7-4273-d382-39775cc3d0b4"
      },
      "source": [
        "ldaseq.print_topics(time=1)"
      ],
      "execution_count": 69,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('covid-19', 0.04660667569843402),\n",
              "  ('pandemic', 0.01656140913626898),\n",
              "  ('de', 0.014408436877613584),\n",
              "  ('la', 0.008500917267636205),\n",
              "  ('impact', 0.006297414325254963),\n",
              "  ('crisis', 0.0048143407065999655),\n",
              "  ('e', 0.004764436467061105),\n",
              "  ('global', 0.004163796548585984),\n",
              "  ('health', 0.004120430568105463),\n",
              "  ('en', 0.004090482350693483),\n",
              "  ('di', 0.003984883267667977),\n",
              "  ('social', 0.003921407729428352),\n",
              "  ('article', 0.0034478729289488847),\n",
              "  ('covid', 0.0034288331732737556),\n",
              "  ('economic', 0.0033368208656752627),\n",
              "  ('time', 0.003329696401301124),\n",
              "  ('medical', 0.003277169388507299),\n",
              "  ('world', 0.003260283503466654),\n",
              "  ('new', 0.003042821886329991),\n",
              "  ('learn', 0.0030156952330389125)],\n",
              " [('covid-19', 0.0638124287465592),\n",
              "  ('coronavirus', 0.026969213165491154),\n",
              "  ('disease', 0.024146529032677923),\n",
              "  ('patient', 0.013237142599127186),\n",
              "  ('pandemic', 0.01224390454015414),\n",
              "  ('sars', 0.01039322933544439),\n",
              "  ('health', 0.01033173903961803),\n",
              "  ('severe', 0.009510829427422055),\n",
              "  ('respiratory', 0.009037157011617634),\n",
              "  ('cov-2', 0.008976071817818567),\n",
              "  ('case', 0.008927837771974173),\n",
              "  ('acute', 0.008156272769640045),\n",
              "  ('china', 0.00807723726407815),\n",
              "  ('cause', 0.007792473477521388),\n",
              "  ('infection', 0.007611490603074289),\n",
              "  ('outbreak', 0.0071779500172514344),\n",
              "  ('covid?\\\\19', 0.006824205845398747),\n",
              "  ('syndrome', 0.006669123360485533),\n",
              "  ('study', 0.006282581951873635),\n",
              "  ('novel', 0.0061415702360409505)]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 69
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "AeSiPOnISpuR",
        "outputId": "2edcbeb8-52dc-4a92-ec91-a9530c916df9"
      },
      "source": [
        "ldaseq.print_topics(time=2)"
      ],
      "execution_count": 70,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('covid-19', 0.046322261608395095),\n",
              "  ('pandemic', 0.016420132572796182),\n",
              "  ('de', 0.01636708598028615),\n",
              "  ('la', 0.009145930256156301),\n",
              "  ('impact', 0.006030995432705853),\n",
              "  ('e', 0.004992934877059675),\n",
              "  ('crisis', 0.004726211224326175),\n",
              "  ('en', 0.004254720878231465),\n",
              "  ('di', 0.004182274110766479),\n",
              "  ('health', 0.003995426408708622),\n",
              "  ('global', 0.003981191806455716),\n",
              "  ('social', 0.003860794888898123),\n",
              "  ('covid', 0.00349350440310812),\n",
              "  ('article', 0.0034025119263356796),\n",
              "  ('time', 0.003223772288278957),\n",
              "  ('economic', 0.0032192136361130076),\n",
              "  ('world', 0.00321826324729171),\n",
              "  ('medical', 0.003173588334713717),\n",
              "  ('y', 0.0031304526048208394),\n",
              "  ('el', 0.0031206825589658238)],\n",
              " [('covid-19', 0.06428464962404917),\n",
              "  ('coronavirus', 0.027254611441029808),\n",
              "  ('disease', 0.02432902692950473),\n",
              "  ('patient', 0.013227578385411444),\n",
              "  ('pandemic', 0.012390558732265263),\n",
              "  ('sars', 0.010489491673089726),\n",
              "  ('health', 0.010097726048698887),\n",
              "  ('respiratory', 0.009111447115459295),\n",
              "  ('cov-2', 0.009055990726999847),\n",
              "  ('case', 0.009021078516874852),\n",
              "  ('severe', 0.008798192169837536),\n",
              "  ('acute', 0.008205517396569361),\n",
              "  ('china', 0.008172925714357054),\n",
              "  ('cause', 0.00787177668801646),\n",
              "  ('infection', 0.007664184090606179),\n",
              "  ('outbreak', 0.007250528971536949),\n",
              "  ('covid?\\\\19', 0.0067780641057692105),\n",
              "  ('syndrome', 0.006722534570332845),\n",
              "  ('study', 0.0062623970194557515),\n",
              "  ('novel', 0.006223719335553649)]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 70
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "BBrbWZkwTjMK",
        "outputId": "b9622515-6ded-4dee-992e-b0aca35671b2"
      },
      "source": [
        "ldaseq.print_topic_times(topic=0) # evolution of 1st topic"
      ],
      "execution_count": 71,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('covid-19', 0.046723443618186804),\n",
              "  ('pandemic', 0.01666185694081015),\n",
              "  ('de', 0.013147735282885771),\n",
              "  ('la', 0.00805385891530807),\n",
              "  ('impact', 0.0064823056876690205),\n",
              "  ('crisis', 0.004873021391956587),\n",
              "  ('e', 0.004615595950721479),\n",
              "  ('global', 0.004246331244818657),\n",
              "  ('health', 0.004170162839912912),\n",
              "  ('social', 0.003980834727989875),\n",
              "  ('en', 0.003974607484036784),\n",
              "  ('di', 0.0038445581715665614),\n",
              "  ('time', 0.0034011534113324473),\n",
              "  ('article', 0.0033993178606789775),\n",
              "  ('economic', 0.0033862981987582675),\n",
              "  ('covid', 0.0033821781033152687),\n",
              "  ('world', 0.0032863856979313067),\n",
              "  ('medical', 0.003206116218973925),\n",
              "  ('new', 0.0030615651854937033),\n",
              "  ('learn', 0.003056385709554595)],\n",
              " [('covid-19', 0.04660667569843402),\n",
              "  ('pandemic', 0.01656140913626898),\n",
              "  ('de', 0.014408436877613584),\n",
              "  ('la', 0.008500917267636205),\n",
              "  ('impact', 0.006297414325254963),\n",
              "  ('crisis', 0.0048143407065999655),\n",
              "  ('e', 0.004764436467061105),\n",
              "  ('global', 0.004163796548585984),\n",
              "  ('health', 0.004120430568105463),\n",
              "  ('en', 0.004090482350693483),\n",
              "  ('di', 0.003984883267667977),\n",
              "  ('social', 0.003921407729428352),\n",
              "  ('article', 0.0034478729289488847),\n",
              "  ('covid', 0.0034288331732737556),\n",
              "  ('economic', 0.0033368208656752627),\n",
              "  ('time', 0.003329696401301124),\n",
              "  ('medical', 0.003277169388507299),\n",
              "  ('world', 0.003260283503466654),\n",
              "  ('new', 0.003042821886329991),\n",
              "  ('learn', 0.0030156952330389125)],\n",
              " [('covid-19', 0.046322261608395095),\n",
              "  ('pandemic', 0.016420132572796182),\n",
              "  ('de', 0.01636708598028615),\n",
              "  ('la', 0.009145930256156301),\n",
              "  ('impact', 0.006030995432705853),\n",
              "  ('e', 0.004992934877059675),\n",
              "  ('crisis', 0.004726211224326175),\n",
              "  ('en', 0.004254720878231465),\n",
              "  ('di', 0.004182274110766479),\n",
              "  ('health', 0.003995426408708622),\n",
              "  ('global', 0.003981191806455716),\n",
              "  ('social', 0.003860794888898123),\n",
              "  ('covid', 0.00349350440310812),\n",
              "  ('article', 0.0034025119263356796),\n",
              "  ('time', 0.003223772288278957),\n",
              "  ('economic', 0.0032192136361130076),\n",
              "  ('world', 0.00321826324729171),\n",
              "  ('medical', 0.003173588334713717),\n",
              "  ('y', 0.0031304526048208394),\n",
              "  ('el', 0.0031206825589658238)]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 71
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/"
        },
        "id": "GX5GVx4rTxMN",
        "outputId": "eeb4059c-7755-4095-f0b0-058102dc3377"
      },
      "source": [
        "ldaseq.print_topic_times(topic=1) # evolution of 2nd topic"
      ],
      "execution_count": 72,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/plain": [
              "[[('covid-19', 0.06346346053327592),\n",
              "  ('coronavirus', 0.02673126389515641),\n",
              "  ('disease', 0.02401511538115626),\n",
              "  ('patient', 0.01330413138984548),\n",
              "  ('pandemic', 0.012128038912016657),\n",
              "  ('sars', 0.010308268450830911),\n",
              "  ('health', 0.00986025848576333),\n",
              "  ('severe', 0.009223478843966168),\n",
              "  ('respiratory', 0.008958171431812089),\n",
              "  ('cov-2', 0.008903325532338982),\n",
              "  ('case', 0.00882721770759777),\n",
              "  ('acute', 0.008104757633943211),\n",
              "  ('china', 0.007967753332214873),\n",
              "  ('cause', 0.007711476302335946),\n",
              "  ('infection', 0.007567542583296424),\n",
              "  ('outbreak', 0.007113328763442282),\n",
              "  ('covid?\\\\19', 0.006904749978605111),\n",
              "  ('syndrome', 0.006592741782246652),\n",
              "  ('study', 0.006369879843431602),\n",
              "  ('clinical', 0.006069589891108472)],\n",
              " [('covid-19', 0.0638124287465592),\n",
              "  ('coronavirus', 0.026969213165491154),\n",
              "  ('disease', 0.024146529032677923),\n",
              "  ('patient', 0.013237142599127186),\n",
              "  ('pandemic', 0.01224390454015414),\n",
              "  ('sars', 0.01039322933544439),\n",
              "  ('health', 0.01033173903961803),\n",
              "  ('severe', 0.009510829427422055),\n",
              "  ('respiratory', 0.009037157011617634),\n",
              "  ('cov-2', 0.008976071817818567),\n",
              "  ('case', 0.008927837771974173),\n",
              "  ('acute', 0.008156272769640045),\n",
              "  ('china', 0.00807723726407815),\n",
              "  ('cause', 0.007792473477521388),\n",
              "  ('infection', 0.007611490603074289),\n",
              "  ('outbreak', 0.0071779500172514344),\n",
              "  ('covid?\\\\19', 0.006824205845398747),\n",
              "  ('syndrome', 0.006669123360485533),\n",
              "  ('study', 0.006282581951873635),\n",
              "  ('novel', 0.0061415702360409505)],\n",
              " [('covid-19', 0.06428464962404917),\n",
              "  ('coronavirus', 0.027254611441029808),\n",
              "  ('disease', 0.02432902692950473),\n",
              "  ('patient', 0.013227578385411444),\n",
              "  ('pandemic', 0.012390558732265263),\n",
              "  ('sars', 0.010489491673089726),\n",
              "  ('health', 0.010097726048698887),\n",
              "  ('respiratory', 0.009111447115459295),\n",
              "  ('cov-2', 0.009055990726999847),\n",
              "  ('case', 0.009021078516874852),\n",
              "  ('severe', 0.008798192169837536),\n",
              "  ('acute', 0.008205517396569361),\n",
              "  ('china', 0.008172925714357054),\n",
              "  ('cause', 0.00787177668801646),\n",
              "  ('infection', 0.007664184090606179),\n",
              "  ('outbreak', 0.007250528971536949),\n",
              "  ('covid?\\\\19', 0.0067780641057692105),\n",
              "  ('syndrome', 0.006722534570332845),\n",
              "  ('study', 0.0062623970194557515),\n",
              "  ('novel', 0.006223719335553649)]]"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 72
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "9jEclOFkUcWr"
      },
      "source": [
        "top_words_topic_0 = {}\n",
        "for i, year in enumerate([2019, 2020, 2021]):\n",
        "    word_lst = ldaseq.print_topics(time=i)[0]\n",
        "    top_words_topic_0[year] = [word for word, prob in word_lst[:10]]"
      ],
      "execution_count": 73,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 355
        },
        "id": "KoFLbQeFUgJG",
        "outputId": "62a86827-cbd0-4803-9259-9788f524eda7"
      },
      "source": [
        "pandas.DataFrame(top_words_topic_0)"
      ],
      "execution_count": 75,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>2019</th>\n",
              "      <th>2020</th>\n",
              "      <th>2021</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>pandemic</td>\n",
              "      <td>pandemic</td>\n",
              "      <td>pandemic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>de</td>\n",
              "      <td>de</td>\n",
              "      <td>de</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>la</td>\n",
              "      <td>la</td>\n",
              "      <td>la</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>impact</td>\n",
              "      <td>impact</td>\n",
              "      <td>impact</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>crisis</td>\n",
              "      <td>crisis</td>\n",
              "      <td>e</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>e</td>\n",
              "      <td>e</td>\n",
              "      <td>crisis</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>global</td>\n",
              "      <td>global</td>\n",
              "      <td>en</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>health</td>\n",
              "      <td>health</td>\n",
              "      <td>di</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>social</td>\n",
              "      <td>en</td>\n",
              "      <td>health</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "       2019      2020      2021\n",
              "0  covid-19  covid-19  covid-19\n",
              "1  pandemic  pandemic  pandemic\n",
              "2        de        de        de\n",
              "3        la        la        la\n",
              "4    impact    impact    impact\n",
              "5    crisis    crisis         e\n",
              "6         e         e    crisis\n",
              "7    global    global        en\n",
              "8    health    health        di\n",
              "9    social        en    health"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 75
        }
      ]
    },
    {
      "cell_type": "code",
      "metadata": {
        "id": "gXBQhU_7UkJJ"
      },
      "source": [
        "top_words_topic_1 = {}\n",
        "for i, year in enumerate([2019, 2020, 2021]):\n",
        "    word_2st = ldaseq.print_topics(time=i)[1]\n",
        "    top_words_topic_1[year] = [word for word, prob in word_2st[:10]]"
      ],
      "execution_count": 76,
      "outputs": []
    },
    {
      "cell_type": "code",
      "metadata": {
        "colab": {
          "base_uri": "https://localhost:8080/",
          "height": 355
        },
        "id": "d0mGPjqSUlUX",
        "outputId": "a53792f5-e232-418e-f84e-0908de4aeee1"
      },
      "source": [
        "pandas.DataFrame(top_words_topic_1)"
      ],
      "execution_count": 77,
      "outputs": [
        {
          "output_type": "execute_result",
          "data": {
            "text/html": [
              "<div>\n",
              "<style scoped>\n",
              "    .dataframe tbody tr th:only-of-type {\n",
              "        vertical-align: middle;\n",
              "    }\n",
              "\n",
              "    .dataframe tbody tr th {\n",
              "        vertical-align: top;\n",
              "    }\n",
              "\n",
              "    .dataframe thead th {\n",
              "        text-align: right;\n",
              "    }\n",
              "</style>\n",
              "<table border=\"1\" class=\"dataframe\">\n",
              "  <thead>\n",
              "    <tr style=\"text-align: right;\">\n",
              "      <th></th>\n",
              "      <th>2019</th>\n",
              "      <th>2020</th>\n",
              "      <th>2021</th>\n",
              "    </tr>\n",
              "  </thead>\n",
              "  <tbody>\n",
              "    <tr>\n",
              "      <th>0</th>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "      <td>covid-19</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>1</th>\n",
              "      <td>coronavirus</td>\n",
              "      <td>coronavirus</td>\n",
              "      <td>coronavirus</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>2</th>\n",
              "      <td>disease</td>\n",
              "      <td>disease</td>\n",
              "      <td>disease</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>3</th>\n",
              "      <td>patient</td>\n",
              "      <td>patient</td>\n",
              "      <td>patient</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>4</th>\n",
              "      <td>pandemic</td>\n",
              "      <td>pandemic</td>\n",
              "      <td>pandemic</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>5</th>\n",
              "      <td>sars</td>\n",
              "      <td>sars</td>\n",
              "      <td>sars</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>6</th>\n",
              "      <td>health</td>\n",
              "      <td>health</td>\n",
              "      <td>health</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>7</th>\n",
              "      <td>severe</td>\n",
              "      <td>severe</td>\n",
              "      <td>respiratory</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>8</th>\n",
              "      <td>respiratory</td>\n",
              "      <td>respiratory</td>\n",
              "      <td>cov-2</td>\n",
              "    </tr>\n",
              "    <tr>\n",
              "      <th>9</th>\n",
              "      <td>cov-2</td>\n",
              "      <td>cov-2</td>\n",
              "      <td>case</td>\n",
              "    </tr>\n",
              "  </tbody>\n",
              "</table>\n",
              "</div>"
            ],
            "text/plain": [
              "          2019         2020         2021\n",
              "0     covid-19     covid-19     covid-19\n",
              "1  coronavirus  coronavirus  coronavirus\n",
              "2      disease      disease      disease\n",
              "3      patient      patient      patient\n",
              "4     pandemic     pandemic     pandemic\n",
              "5         sars         sars         sars\n",
              "6       health       health       health\n",
              "7       severe       severe  respiratory\n",
              "8  respiratory  respiratory        cov-2\n",
              "9        cov-2        cov-2         case"
            ]
          },
          "metadata": {
            "tags": []
          },
          "execution_count": 77
        }
      ]
    }
  ]
}