lexical_model.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from nltk.tokenize import word_tokenize\n",
    "import matplotlib.pyplot as plt \n",
    "import re\n",
    "import demoji\n",
    "# demoji.download_codes()\n",
    "import nltk\n",
    "from nltk.corpus import stopwords\n",
    "from wordcloud import WordCloud,STOPWORDS\n",
    "import seaborn as sns\n",
    "\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'0.1.5'"
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "demoji.__version__"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Lexical Model"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Downloading package punkt to /Users/anweshcr7/nltk_data...\n",
      "[nltk_data]   Package punkt is already up-to-date!\n",
      "[nltk_data] Downloading package wordnet to\n",
      "[nltk_data]     /Users/anweshcr7/nltk_data...\n",
      "[nltk_data]   Package wordnet is already up-to-date!\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "True"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# conda install nltk\n",
    "nltk.download('punkt')\n",
    "nltk.download('wordnet')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Parse data from file"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [],
   "source": [
    "def parse_dataset(fp):\n",
    "    '''\n",
    "    Loads the dataset file with label-tweet on each line and parses the dataset.\n",
    "    :param fp: filepath of dataset\n",
    "    :return:\n",
    "        corpus: list of tweet strings of each tweet.\n",
    "        y: list of labels\n",
    "    '''\n",
    "    y = []\n",
    "    corpus = []\n",
    "    with open(fp, 'rt') as data_in:\n",
    "        for line in data_in:\n",
    "            if not line.lower().startswith(\"tweet index\"): # discard first line if it contains metadata\n",
    "                line = line.rstrip() # remove trailing whitespace\n",
    "                label = int(line.split(\"\\t\")[1])\n",
    "                tweet = line.split(\"\\t\")[2]\n",
    "                y.append(label)\n",
    "                corpus.append(tweet)\n",
    "\n",
    "    return corpus, y"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Train Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "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>tweet</th>\n",
       "      <th>label</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>Sweet United Nations video. Just in time for C...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>@mrdahl87 We are rumored to have talked to Erv...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>Hey there! Nice to see you Minnesota/ND Winter...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>3 episodes left I'm dying over here</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>\"I can't breathe!\" was chosen as the most nota...</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               tweet  label\n",
       "0  Sweet United Nations video. Just in time for C...      1\n",
       "1  @mrdahl87 We are rumored to have talked to Erv...      1\n",
       "2  Hey there! Nice to see you Minnesota/ND Winter...      1\n",
       "3                3 episodes left I'm dying over here      0\n",
       "4  \"I can't breathe!\" was chosen as the most nota...      1"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data, y_label = parse_dataset('SemEval2018-T3-train-taskA_emoji.txt')\n",
    "data_df = pd.DataFrame(np.array(data).reshape(3834,1), columns = ['tweet'])\n",
    "data_df['label'] = np.array(y_label).reshape(3834,1)\n",
    "data_df.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Test Data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 784 entries, 0 to 783\n",
      "Data columns (total 2 columns):\n",
      "tweet    784 non-null object\n",
      "label    784 non-null int64\n",
      "dtypes: int64(1), object(1)\n",
      "memory usage: 12.4+ KB\n"
     ]
    }
   ],
   "source": [
    "# Test the model\n",
    "test_data, test_label = parse_dataset('SemEval2018-T3_gold_test_taskA_emoji.txt')\n",
    "test_df = pd.DataFrame(np.array(test_data).reshape(784,1), columns = ['tweet'])\n",
    "test_df['label'] = np.array(test_label).reshape(784,1)\n",
    "test_df.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "ename": "NameError",
     "evalue": "name 'lexical_df' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-7-7eec3a5ea49f>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m     23\u001b[0m \u001b[0;31m# re.findall(r'http\\S+', trial)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m     24\u001b[0m \u001b[0;31m# re.findall(r'([.,/#!$%^&*;:{}=_`~()-])[.,/#!$%^&*;:{}=_`~()-]+', '!# dfre *')\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m---> 25\u001b[0;31m \u001b[0mlexical_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0mlexical_df\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'punc_flooding'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m==\u001b[0m \u001b[0;36m1\u001b[0m\u001b[0;34m]\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m     26\u001b[0m \u001b[0;31m# trial.apply(lambda x: len(list(ch for ch in x if ch.isupper() == 1)))\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mNameError\u001b[0m: name 'lexical_df' is not defined"
     ]
    }
   ],
   "source": [
    "# # data_df.info()\n",
    "trial = data_df['tweet'][2]\n",
    "# re.findall(r'\\B@\\w*[a-zA-Z]+\\w*', trial)\n",
    "# # for idx, row in data_df.iterrows():\n",
    "# #     if(demoji.findall(row['tweet'])):\n",
    "# #         print(idx)\n",
    "# #         break\n",
    "    \n",
    "# len(demoji.findall(trial))\n",
    "# len(trial.split(' '))\n",
    "# import string\n",
    "\n",
    "# count_punc = lambda l1, l2: len(list(filter(lambda c: c in l2, l1)))\n",
    "\n",
    "# string.punctuation\n",
    "# count_punc(trial, string.punctuation)\n",
    "# data_df[['tweet','label']].apply(lambda row: print(row))\n",
    "# gen = list(ch for ch in trial if ch.isupper() == 1)\n",
    "# gen\n",
    "# for x in gen:\n",
    "#     print(x)\n",
    "# re.findall(r'(http://[^\\s]+)', trial)\n",
    "# re.findall(r'http\\S+', trial)\n",
    "# re.findall(r'([.,/#!$%^&*;:{}=_`~()-])[.,/#!$%^&*;:{}=_`~()-]+', '!# dfre *')\n",
    "lexical_df[lexical_df['punc_flooding'] == 1]\n",
    "# trial.apply(lambda x: len(list(ch for ch in x if ch.isupper() == 1)))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "import string\n",
    "count_punc = lambda l1, l2: len(list(filter(lambda c: c in l2, l1)))\n",
    "\n",
    "def lexical_feature_extract(og_df):\n",
    "    df = og_df.copy()\n",
    "    df['capitalisation'] = df['tweet'].apply(lambda x: len(list(ch for ch in x if ch.isupper() == 1))/len(x))\n",
    "    df['url'] = df['tweet'].apply(lambda x: 1 if len(re.findall(r'http\\S+', x)) else 0)\n",
    "    df['hashtag_counts'] = df['tweet'].apply(lambda x: len(re.findall(r'\\B#\\w*[a-zA-Z]+\\w*', x)))\n",
    "#     df['emoji_counts'] = df['tweet'].apply(lambda x: len(demoji.findall(row['tweet'])))\n",
    "    df['@_counts'] = df['tweet'].apply(lambda x: len(re.findall(r'\\B@\\w*[a-zA-Z]+\\w*', x)))\n",
    "    df['length'] = df['tweet'].apply(lambda x: len(x.split(' ')))\n",
    "    df['hashtag_word_ratio'] = df['hashtag_counts']/df['length']\n",
    "    df['# punctuation'] = df['tweet'].apply(lambda x: count_punc(x, string.punctuation))\n",
    "#     df['tweet_label'] = df['label'].apply(lambda x: 'Irony' if x else 'Non-Irony')\n",
    "    df['punc_flooding'] = df['tweet'].apply(lambda x: 1 if len(re.findall(r'([.,/#!$%^&*;:{}=_`~()-])[.,/#!$%^&*;:{}=_`~()-]+', x)) else 0)\n",
    "    return df.drop(['tweet'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "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>label</th>\n",
       "      <th>capitalisation</th>\n",
       "      <th>url</th>\n",
       "      <th>hashtag_counts</th>\n",
       "      <th>@_counts</th>\n",
       "      <th>length</th>\n",
       "      <th>hashtag_word_ratio</th>\n",
       "      <th># punctuation</th>\n",
       "      <th>punc_flooding</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0.110000</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>13</td>\n",
       "      <td>0.153846</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0.039683</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>24</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>11</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>2</td>\n",
       "      <td>1</td>\n",
       "      <td>0.129630</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>0.028571</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>7</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>4</td>\n",
       "      <td>1</td>\n",
       "      <td>0.024000</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>23</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3829</td>\n",
       "      <td>0</td>\n",
       "      <td>0.061224</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>7</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3830</td>\n",
       "      <td>0</td>\n",
       "      <td>0.030769</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>12</td>\n",
       "      <td>0.000000</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3831</td>\n",
       "      <td>0</td>\n",
       "      <td>0.029412</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>19</td>\n",
       "      <td>0.052632</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3832</td>\n",
       "      <td>0</td>\n",
       "      <td>0.095588</td>\n",
       "      <td>1</td>\n",
       "      <td>8</td>\n",
       "      <td>0</td>\n",
       "      <td>18</td>\n",
       "      <td>0.444444</td>\n",
       "      <td>21</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <td>3833</td>\n",
       "      <td>0</td>\n",
       "      <td>0.039216</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>21</td>\n",
       "      <td>0.047619</td>\n",
       "      <td>5</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3834 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "      label  capitalisation  url  hashtag_counts  @_counts  length  \\\n",
       "0         1        0.110000    1               2         0      13   \n",
       "1         1        0.039683    0               0         1      24   \n",
       "2         1        0.129630    0               0         0       9   \n",
       "3         0        0.028571    0               0         0       7   \n",
       "4         1        0.024000    0               0         0      23   \n",
       "...     ...             ...  ...             ...       ...     ...   \n",
       "3829      0        0.061224    0               0         1       7   \n",
       "3830      0        0.030769    0               0         1      12   \n",
       "3831      0        0.029412    0               1         0      19   \n",
       "3832      0        0.095588    1               8         0      18   \n",
       "3833      0        0.039216    0               1         0      21   \n",
       "\n",
       "      hashtag_word_ratio  # punctuation  punc_flooding  \n",
       "0               0.153846              9              1  \n",
       "1               0.000000             11              1  \n",
       "2               0.000000              2              0  \n",
       "3               0.000000              1              0  \n",
       "4               0.000000              4              0  \n",
       "...                  ...            ...            ...  \n",
       "3829            0.000000              1              0  \n",
       "3830            0.000000              2              0  \n",
       "3831            0.052632              4              0  \n",
       "3832            0.444444             21              1  \n",
       "3833            0.047619              5              1  \n",
       "\n",
       "[3834 rows x 9 columns]"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lexical_df = lexical_feature_extract(data_df)\n",
    "lexical_df\n",
    "# lexical_df[lexical_df['hashtag_counts'] < 1]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Histogram of the the lexical features"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array([[<matplotlib.axes._subplots.AxesSubplot object at 0x1a2708d250>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1a278e1fd0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1a279155d0>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x1a27941ed0>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1a27972810>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1a27972850>],\n",
       "       [<matplotlib.axes._subplots.AxesSubplot object at 0x1a279a4190>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1a27a00410>,\n",
       "        <matplotlib.axes._subplots.AxesSubplot object at 0x1a27a2bd10>]],\n",
       "      dtype=object)"
      ]
     },
     "execution_count": 22,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1152x1440 with 9 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "lexical_df.hist(figsize=(16, 20), bins=50, xlabelsize=8, ylabelsize=8)\n",
    "# lexical_df['punc_count'].describe()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0    1923\n",
       "1    1911\n",
       "Name: label, dtype: int64"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "lexical_df['label'].value_counts()\n",
    "# lexical_df[lexical_df['label']==1]['length']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Punctuation Histogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "ename": "ValueError",
     "evalue": "Could not interpret input 'tweet_label'",
     "output_type": "error",
     "traceback": [
      "\u001b[0;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[0;31mValueError\u001b[0m                                Traceback (most recent call last)",
      "\u001b[0;32m<ipython-input-24-51f8dbafff7b>\u001b[0m in \u001b[0;36m<module>\u001b[0;34m\u001b[0m\n\u001b[1;32m      1\u001b[0m \u001b[0mplt\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mrcParams\u001b[0m\u001b[0;34m[\u001b[0m\u001b[0;34m'figure.figsize'\u001b[0m\u001b[0;34m]\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m(\u001b[0m\u001b[0;36m8\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0;36m6\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m----> 2\u001b[0;31m \u001b[0msns\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mcountplot\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mx\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'# punctuation'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mhue\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m'tweet_label'\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mdata\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0mlexical_df\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpalette\u001b[0m\u001b[0;34m=\u001b[0m\u001b[0;34m\"Set2\"\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m;\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/seaborn/categorical.py\u001b[0m in \u001b[0;36mcountplot\u001b[0;34m(x, y, hue, data, order, hue_order, orient, color, palette, saturation, dodge, ax, **kwargs)\u001b[0m\n\u001b[1;32m   3551\u001b[0m                           \u001b[0mestimator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mci\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_boot\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0munits\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3552\u001b[0m                           \u001b[0morient\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mcolor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpalette\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msaturation\u001b[0m\u001b[0;34m,\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m-> 3553\u001b[0;31m                           errcolor, errwidth, capsize, dodge)\n\u001b[0m\u001b[1;32m   3554\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   3555\u001b[0m     \u001b[0mplotter\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mvalue_label\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"count\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/seaborn/categorical.py\u001b[0m in \u001b[0;36m__init__\u001b[0;34m(self, x, y, hue, data, order, hue_order, estimator, ci, n_boot, units, orient, color, palette, saturation, errcolor, errwidth, capsize, dodge)\u001b[0m\n\u001b[1;32m   1605\u001b[0m         \u001b[0;34m\"\"\"Initialize the plotter.\"\"\"\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1606\u001b[0m         self.establish_variables(x, y, hue, data, orient,\n\u001b[0;32m-> 1607\u001b[0;31m                                  order, hue_order, units)\n\u001b[0m\u001b[1;32m   1608\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mestablish_colors\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mcolor\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mpalette\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0msaturation\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m   1609\u001b[0m         \u001b[0mself\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mestimate_statistic\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0mestimator\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mci\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mn_boot\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;32m~/anaconda3/lib/python3.7/site-packages/seaborn/categorical.py\u001b[0m in \u001b[0;36mestablish_variables\u001b[0;34m(self, x, y, hue, data, orient, order, hue_order, units)\u001b[0m\n\u001b[1;32m    153\u001b[0m                 \u001b[0;32mif\u001b[0m \u001b[0misinstance\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m,\u001b[0m \u001b[0mstring_types\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m:\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    154\u001b[0m                     \u001b[0merr\u001b[0m \u001b[0;34m=\u001b[0m \u001b[0;34m\"Could not interpret input '{}'\"\u001b[0m\u001b[0;34m.\u001b[0m\u001b[0mformat\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0minput\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0;32m--> 155\u001b[0;31m                     \u001b[0;32mraise\u001b[0m \u001b[0mValueError\u001b[0m\u001b[0;34m(\u001b[0m\u001b[0merr\u001b[0m\u001b[0;34m)\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n\u001b[0m\u001b[1;32m    156\u001b[0m \u001b[0;34m\u001b[0m\u001b[0m\n\u001b[1;32m    157\u001b[0m             \u001b[0;31m# Figure out the plotting orientation\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0;34m\u001b[0m\u001b[0m\n",
      "\u001b[0;31mValueError\u001b[0m: Could not interpret input 'tweet_label'"
     ]
    }
   ],
   "source": [
    "plt.rcParams['figure.figsize'] = (8, 6)\n",
    "sns.countplot(x='# punctuation', hue='tweet_label', data=lexical_df, palette=\"Set2\");"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Hashtag-to-word Ratio Histogram"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 692,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 1 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 372,
       "width": 551
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%config InlineBackend.figure_format ='retina'\n",
    "\n",
    "plt.rcParams['figure.figsize'] = (8, 6)\n",
    "sns.countplot(x='hashtag_word_ratio', hue='tweet_label', data=lexical_df[lexical_df['hashtag_word_ratio']>0.4], palette=\"Set2\");\n",
    "\n",
    "# sns.distplot(lexical_df[lexical_df['label']==0]['length'], bins=20, kde=False, rug=True);"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 683,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sns.pairplot(lexical_df)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Prepare train and test df for classification"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [],
   "source": [
    "y_train = lexical_df['label']\n",
    "lexical_df = lexical_df.drop(['label'], axis=1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [],
   "source": [
    "test_lex_df = lexical_feature_extract(test_df)\n",
    "y_test = test_lex_df['label']\n",
    "test_lex_df = test_lex_df.drop(['label'], axis = 1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Random Forest Classifier"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CROSS VALIDATION\n",
      "Score: [0.55584416 0.53645833 0.6015625  0.61096606 0.5691906  0.60835509\n",
      " 0.5770235  0.60835509 0.57441253 0.59791123]\n",
      "Mean: 0.58400790883942\n",
      "std: 0.02414196195756088\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0       0.72      0.68      0.70       473\n",
      "           1       0.55      0.60      0.57       311\n",
      "\n",
      "    accuracy                           0.65       784\n",
      "   macro avg       0.64      0.64      0.64       784\n",
      "weighted avg       0.65      0.65      0.65       784\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# cross-validation on train\n",
    "from sklearn.ensemble import RandomForestClassifier\n",
    "from sklearn import metrics\n",
    "from sklearn.model_selection import cross_val_score\n",
    "\n",
    "# Example of cross-validation\n",
    "rf = RandomForestClassifier(n_estimators=10)\n",
    "\n",
    "score = cross_val_score(rf, lexical_df, y_train, cv=10, scoring = \"accuracy\")\n",
    "print('CROSS VALIDATION')\n",
    "print(\"Score:\", score)\n",
    "print(\"Mean:\", score.mean())\n",
    "print(\"std:\", score.std())\n",
    "\n",
    "rf.fit(lexical_df, y_train)\n",
    "y_pred = rf.predict(test_lex_df)\n",
    "print(metrics.classification_report(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "('capitalisation', 0.40343463696456394)\n",
      "('url', 0.022888925640700684)\n",
      "('hashtag_counts', 0.03786670696567614)\n",
      "('@_counts', 0.05858229688075124)\n",
      "('length', 0.23315010802648733)\n",
      "('hashtag_word_ratio', 0.08258975370468524)\n",
      "('# punctuation', 0.14539454281029524)\n",
      "('punc_flooding', 0.016093029006840193)\n"
     ]
    }
   ],
   "source": [
    "feat_labels = list(lexical_df)\n",
    "for feature in zip(feat_labels, rf.feature_importances_):\n",
    "    print(feature)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Confusion Matrix"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdMAAAGDCAYAAABwcPpaAAAABHNCSVQICAgIfAhkiAAAAAlwSFlzAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMi40LCBodHRwOi8vbWF0cGxvdGxpYi5vcmcv7US4rQAAIABJREFUeJzt3XecJUW5//HPd3eJAqICShRUMCFZ5KIIKqIoCmaUJKh7VUwXEVExYI6oGH8YUBAVFPBiAhUjXPIKKKKERQRZcgYFduf5/XF6YVhnZ4bt7T0zZz5vXv3ac6r7VNWM4zzzVFVXp6qQJEmLblq/OyBJ0mRnMJUkqSWDqSRJLRlMJUlqyWAqSVJLBlNJkloymGpKSrJckh8nuSXJD1rUs1uSXyzOvvVDkp8n2avf/ZAmK4OpJrQkr0pydpLbk8xpfuk/bTFU/VLg4cDDqupli1pJVR1VVTsshv7cT5LtklSS4xYo37gp/+046/lAku+MdV1V7VhV317E7kpTnsFUE1aS/YDPAR+lF/jWAb4M7LwYqn8kcFFVzV0MdXXlOmDrJA8bVrYXcNHiaiA9/h6QWvL/RJqQkjwY+CCwb1UdV1V3VNU9VfXjqnpHc80yST6X5Krm+FySZZpz2yW5Msnbk1zbZLV7N+cOBt4HvKLJeF+zYAaXZN0mA5zRvH91ktlJbktyWZLdhpWfMuxzWyc5qxk+PivJ1sPO/TbJh5Kc2tTziySrjPJtuBv4EbBr8/npwMuBoxb4Xn0+yRVJbk1yTpJtmvLnAu8e9nWeN6wfH0lyKnAn8Kim7LXN+a8k+eGw+j+R5OQkGff/gNIUYzDVRPVfwLLA8aNc8x5gK2ATYGNgS+CgYecfATwYWBN4DfClJA+pqvfTy3aPrqoVquobo3UkyYOAQ4Edq2pFYGvg3BGueyjw0+bahwGHAD9dILN8FbA3sBqwNLD/aG0DRwB7Nq+fA1wAXLXANWfR+x48FPgu8IMky1bViQt8nRsP+8wewExgReDyBep7O7BR84fCNvS+d3uVe49KC2Uw1UT1MOD6MYZhdwM+WFXXVtV1wMH0gsR89zTn76mqnwG3A49dxP4MARsmWa6q5lTVBSNc83zg4qo6sqrmVtX3gL8CLxh2zeFVdVFV/Qs4hl4QXKiq+j/goUkeSy+oHjHCNd+pqhuaNj8DLMPYX+e3quqC5jP3LFDfncDu9P4Y+A7w5qq6coz6pCnNYKqJ6gZglfnDrAuxBvfPqi5vyu6tY4FgfCewwgPtSFXdAbwCeD0wJ8lPkzxuHP2Z36c1h72/ehH6cyTwJuAZjJCpN0PZFzZDyzfTy8ZHGz4GuGK0k1V1JjAbCL2gL2kUBlNNVKcB/wZ2GeWaq+gtJJpvHf5zCHS87gCWH/b+EcNPVtVJVfVsYHV62ebXxtGf+X365yL2ab4jgTcCP2uyxns1w7DvpDeX+pCqWhm4hV4QBFjY0OyoQ7ZJ9qWX4V4FHLDoXZemBoOpJqSquoXeIqEvJdklyfJJlkqyY5JPNpd9DzgoyarNQp730RuWXBTnAk9Psk6z+Old808keXiSFzZzp3fRGy6eN0IdPwM2aG7nmZHkFcATgJ8sYp8AqKrLgG3pzREvaEVgLr2VvzOSvA9Yadj5a4B1H8iK3SQbAB+mN9S7B3BAklGHo6WpzmCqCauqDgH2o7eo6Dp6Q5NvorfCFXq/8M8Gzgf+BMxqyhalrV8CRzd1ncP9A+A0eotyrgJupBfY3jhCHTcAOzXX3kAvo9upqq5flD4tUPcpVTVS1n0S8HN6t8tcTi+bHz6EO39DihuSzBqrnWZY/TvAJ6rqvKq6mN6K4CPnr5SW9J/iAj1JktoxM5UkqSWDqSRJLRlMJUlqyWAqSVJLBlNJkloabXeZvrrn+tkuM9akt+WGe4x9kTQJ/PHqUzt70EHb3/dLrfKovj+EwcxUktRfQ/PaHWNIsmySM5Ocl+SC5slRJDkqyd+S/DnJN5Ms1ZQnyaFJLklyfpLNxmrDYCpJGnR3Ac9snpy0CfDcJFvRe5zh44AnAcsBr22u3xFYvzlmAl8Zq4EJO8wrSZoiaqjb6nu7E93evF2qOap5mhQASc4E1mre7gwc0Xzu9CQrJ1m9quYsrA0zU0lSfw0NtTvGIcn0JOcC1wK/rKozhp1bit4+1Cc2RWty/205r+T+T3/6DwZTSVJfVQ21OpLMTHL2sGPmf7ZR86pqE3rZ55ZJNhx2+svA76vqD837kRY0jbpIymFeSdKkVlWHAYeN89qbk/wWeC7w5yTvB1YF/nvYZVcCaw97vxZjPN7RzFSS1F8dD/M2j2lcuXm9HLA98NckrwWeA7yy6n4TtycAezarercCbhltvhTMTCVJ/dbxAiRgdeDbSabTSyKPqaqfJJlL79GFpyUBOK6qPkjv2cTPAy4B7gT2HqsBg6kkqb/Gca9oG1V1PrDpCOUjxsBmFe++D6QNg6kkqb+6z0w755ypJEktmZlKkvprnPeKTmQGU0lSX9UADPMaTCVJ/WVmKklSSwOQmboASZKklsxMJUn91fF9pkuCwVSS1F8DMMxrMJUk9dcALEByzlSSpJbMTCVJ/eUwryRJLQ3AMK/BVJLUV1Wu5pUkqZ0BGOZ1AZIkSS2ZmUqS+ss5U0mSWhqAYV6DqSSpv9xOUJKklgYgM3UBkiRJLZmZSpL6ywVIkiS1NADDvAZTSVJ/DUBm6pypJEktmZlKkvprADJTg6kkqa/c6F6SpLbMTCVJamkAVvO6AEmSpJbMTCVJ/eUwryRJLQ3AMK/BVJLUX2amkiS1NACZqQuQJElqycxUktRfDvNKktSSwVSSpJacM5UkSWamkqT+cphXkqSWBmCY12AqSeovM1NJkloagMzUBUiSJLVkZipJ6i+HeSVJaslgKklSS1X97kFrBlNJUn8NQGbqAiRJkloyM5Uk9dcAZKYGU0lSfw3AfaYGU0lSfw1AZuqcqSRJLZmZSpL6y1tjJElqaQCGeQ2mkqT+MphKktTSAKzmdQGSJGmgJVk7yW+SXJjkgiRvbco3SXJ6knOTnJ1ky6Y8SQ5NckmS85NsNlYbZqaSpL6qoc4XIM0F3l5Vs5KsCJyT5JfAJ4GDq+rnSZ7XvN8O2BFYvzmeAnyl+XehDKaSpP7qeM60quYAc5rXtyW5EFgTKGCl5rIHA1c1r3cGjqiqAk5PsnKS1Zt6RmQwlST1V8s50yQzgZnDig6rqsMWcu26wKbAGcDbgJOSfJretOfWzWVrAlcM+9iVTZnBVJI0QbUc5m0C54jBc7gkKwDHAm+rqluTfBj4n6o6NsnLgW8A2wMZqZnR6nYBkiRp4CVZil4gPaqqjmuK9wLmv/4BsGXz+kpg7WEfX4v7hoBHZDCVJPXX0FC7YwxJQi/rvLCqDhl26ipg2+b1M4GLm9cnAHs2q3q3Am4Zbb4UHOaVJPVb95s2PBXYA/hTknObsncDrwM+n2QG8G/um3f9GfA84BLgTmDvsRowmEqS+qvjvXmr6hRGngcF2HyE6wvY94G04TCvJEktmZkOsLvuupu99n0Hd99zD/PmzuPZz3gab3rtHrzzA5/ggr9ezIwZM9jwCRvw/gPewlIzZvDrP5zGF752BNMyjenTp3PgW2ey2cYb9vvLkHj/Z9/F05/9VG68/iZett0eAPz3/vvw4t1eyE033AzAFz/2/zjl5NMA2OfNe7Dzq3ZiaN4Qnzzos5z22zP71neNg3vzaiJbeuml+OahH2f55Zfjnrlz2fMN+7PNVlvw/B2ewcfffwAAB3zgExz74xPZ9UU7sdXmm/CMp21FEv52yWXs/96P8uPvfa3PX4UEPz76Zxz9zWP50Bfee7/y7xx2NEd+5Xv3K3vUBuvynF2exUu33Z1VH7EKXz3m8+yy9a4MDcAv7IHV/Q5InXOYd4AlYfnllwNg7ty5zJ07lyQ8festSUISnvT4x3LNtdcDsPzyy9Fb9Ab/+ve/IQubYpCWrFmnn8ctN986rmu3e842nPSjk7nn7nu46h9zuOKyK9lw08d33EO1UkPtjgmg02DabBy8b5KHdNmOFm7evHm8ZK99efpOr+S/nrwpGz3xcfeeu2fuXH580sk87Slb3Fv2q9+dygte+TreuP/7+NC7/6cfXZbGbdd9XsLRv/427//su1jxwSsCsOrqq3L1Vdfce821c65ltdVX7VcXNR5D1e6YALrOTHcF1gDOSvL9JM9JFp7uJJnZBOCzv37E9xZ2mR6A6dOnc+y3v8TJxx/Jn/5yERfP/vu95z786S+x+cYbsvkm982Lbr/tU/nx977GoR9/H1/82hF96LE0Pj/41vG84CkvZ9dnvZrrr7mB/T7wJmDkAZWOF4tK3QbTqrqkqt4DbAB8F/gm8I8kByd56AjXH1ZVW1TVFq/d85Vddm3KWWnFFXjyZhtxyulnA/Dlbx7FTTffwgFvmTni9Vts8iSu+Occbrr5liXZTWncbrz+JoaGhqgqjjvqBDbc9AkAXHvVdTxijYffe91qq6/GdVdf169uahxqaKjVMRF0PmeaZCPgM8Cn6G3l9FLgVuDXXbc91d14083cetvtAPz7rrs4/aw/st4j1+aHJ5zIqWecwycPfifTpt33I/CPK6+imj/h//K3S7jnnrms/OCVRqxb6rdVVnvYva+fueO2XPrX2QD89hen8JxdnsVSSy/FGuuszjqPWos///HCfnVT4zEAw7ydruZNcg5wM71tnA6sqruaU2ckeWqXbQuuu+Em3vPhTzNvaIgaKp7zzG3Y7qlPYeOnP5/VH74au83cD4Dtt92aN+yzG7/87Smc8POTmTFjBssuszSf/uCBjDIqLy0xH/vKB9h8601Z+aErc+Ks4/nqp77B5ltvymM3XJ+qYs4VV/Phd3wSgNl/u4xfnPBrjv39UcybO4+Pv+sQV/JOdBNkEVEbqQ4nE5I8qqpmL8pn77l+9sT4c0NqYcsN9+h3F6TF4o9Xn9rZX9Z3fHj3Vr/vH3TQd/r+V3/X95n+M8mrgHWHt1VVH+y4XUnSZDFBhmrb6DqY/i9wC3AOcNcY10qSpqIBGIbvOpiuVVXP7bgNSdJkNgCZaderef8vyZM6bkOSNJkNwA5IXWemTwNeneQyesO8ofd0m406bleSpCWm62C6Y8f1S5ImuwEY5u00mFbV5Uk2BrZpiv5QVed12aYkaXKZKLsYtdH1RvdvBY4CVmuO7yR5c5dtSpImGXdAGtNrgKdU1R0AST4BnAZ8oeN2JUmTxQQJiG10vZo3wLxh7+c1ZZIkDYyuM9PD6e3De3zzfhd6+/RKktQzQW5vaaPrBUiHJPktvVtkAuxdVX/ssk1J0iQzAMO8nQXTJNOA86tqQ2BWV+1Ikia3GoBg2tmcaVUNAeclWaerNiRJmgi6njNdHbggyZnAHfMLq+qFHbcrSZosBiAz7TqYHtxx/ZKkyW4ANm3oegHS77qsX5I0AMxMR5bkNmCk7878je5X6qJdSdIkZDAdWVWt2EW9kiRNRF3PmUqSNKoqM1NJktpxmFeSpJYMppIkteMOSJIkycxUktRnA5CZGkwlSf01+TdAMphKkvrLOVNJkmRmKknqswHITA2mkqT+cs5UkqR2BmHO1GAqSeqvAchMXYAkSVJLZqaSpL5ymFeSpLYGYJjXYCpJ6qsymEqS1NIABFMXIEmS1JKZqSSprxzmlSSpLYOpJEntDEJm6pypJEktmZlKkvpqEDJTg6kkqa8MppIktVXpdw9aM5hKkvpqEDJTFyBJktSSwVSS1Fc1lFbHWJKsneQ3SS5MckGSty5wfv8klWSV5n2SHJrkkiTnJ9lsrDYc5pUk9dUSGOadC7y9qmYlWRE4J8kvq+ovSdYGng38Y9j1OwLrN8dTgK80/y6Umakkqa+q0uoYu/6aU1Wzmte3ARcCazanPwscAAx/qOrOwBHVczqwcpLVR2vDYCpJ6qsaanckmZnk7GHHzIW1lWRdYFPgjCQvBP5ZVectcNmawBXD3l/JfcF3RGMO8yZ5MfDLqrotyYHAZsBHq+rcsT4rSVLXquow4LCxrkuyAnAs8DZ6Q7/vAXYY6dKRmhmt7vFkph9oAunWwAuAo4GvjuNzkiSNqesFSABJlqIXSI+qquOARwPrAecl+TuwFjArySPoZaJrD/v4WsBVo9U/nmA6r/l3J+DLVXUssMy4ei9J0hiq2h1jSRLgG8CFVXVIr836U1WtVlXrVtW69ALoZlV1NXACsGezqncr4JaqmjNaG+NZzTsnyZeA5wJbJFka51olSYvJeLPLFp4K7AH8Kcn8Kcp3V9XPFnL9z4DnAZcAdwJ7j9XAeILpy5tKv1BVNyVZAzhwHJ+TJKnvquoURp4HHX7NusNeF7DvA2ljocE0yUrD3p44rOx24NQH0ogkSQuzBDLTzo2WmV5Ab/XS8K9y/vsC1umwX5KkKWI8854T3UKDaVWtvbBzkiQtLoOQmY5rIVGSXZO8u3m9VpLNu+2WJGmq6HoHpCVhzGCa5IvAM+ithILeyibvM5UkqTGe1bxbV9VmSf4IUFU3NrfHSJLU2iA8z3Q8wfSeJNNotlJK8jBgAL50SdJEMDRBhmrbGE8w/RK9LZhWTXIwvftOD+60V5KkKWOizHu2MWYwraojkpwDbN8Uvayq/txttyRJU8UgrOYd78PBpwP30BvqdStBSZKGGc9q3vcA3wPWoLdz/neTvKvrjkmSpoauN7pfEsaTme4ObF5VdwIk+QhwDvCxLjsmSZoapsow7+ULXDcDmN1NdyRJU81Ar+ZN8ll6c6R3AhckOal5vwNwypLpniRJE99omen8FbsXAD8dVn56d92RJE01A31rTFV9Y0l2RJI0NU2URURtjDlnmuTRwEeAJwDLzi+vqg067JckaYoYhDnT8dwz+i3gcHrPMd0ROAb4fod9kiRNIVPiqTHA8lV1EkBVXVpVB9F7iowkSWJ8t8bclSTApUleD/wTWK3bbkmSpoopMWcK/A+wAvAWenOnDwb26bJTAMutsU3XTUide8ca2/a7C9KENwhzpuPZ6P6M5uVt3PeAcEmSFouJMu/ZxmibNhxP8wzTkVTVizvpkSRpShn0zPSLS6wXkiRNYqNt2nDykuyIJGlqGoD1R+N+nqkkSZ0Y9GFeSZI6NwgLkMazaQMASZbpsiOSJE1WYwbTJFsm+RNwcfN+4yRf6LxnkqQpYajlMRGMJzM9FNgJuAGgqs7D7QQlSYtJkVbHRDCeOdNpVXV5b0fBe83rqD+SpClmaACW844nmF6RZEugkkwH3gxc1G23JElTxdAEyS7bGM8w7xuA/YB1gGuArZoySZLE+PbmvRbYdQn0RZI0BU2Uec82xgymSb7GCBtUVNXMTnokSZpSJsqK3DbGM2f6q2GvlwVeBFzRTXckSVPNlMhMq+ro4e+THAn8srMeSZI0ySzKdoLrAY9c3B2RJE1NU2KYN8lN3DdnOg24ETiwy05JkqaOgQ+m6e3UsDHwz6ZoqKoG4PZaSdJEMfBzplVVSY6vqs2XVIckSVPL0OSPpePatOHMJJt13hNJkiaphWamSWZU1VzgacDrklwK3AGEXtJqgJUktTYI2wmONsx7JrAZsMsS6oskaQoahIU4owXTAFTVpUuoL5KkKWjQV/OummS/hZ2sqkM66I8kaYoZymAP804HVoABGMyWJKlDowXTOVX1wSXWE0nSlDQl5kwlSerSoM+ZPmuJ9UKSNGUN9KYNVXXjkuyIJEmT1aI8NUaSpMVm0DdtkCSpc4O+AEmSpM4NwpypwVSS1FeDsJp3PE+NkSRJozAzlST11SDMmZqZSpL6aijtjrEkWTvJb5JcmOSCJG9tyh+a5JdJLm7+fUhTniSHJrkkyfnjeaa3wVSS1FdDLY9xmAu8vaoeD2wF7JvkCcCBwMlVtT5wcvMeYEdg/eaYCXxlrAYMppKkvuo6mFbVnKqa1by+DbgQWBPYGfh2c9m3ue/53TsDR1TP6cDKSVYfrQ2DqSRpUksyM8nZw46Zo1y7LrApcAbw8KqaA72AC6zWXLYmcMWwj13ZlC2UC5AkSX1VLe8zrarDgMPGui7JCsCxwNuq6tYs/DmqI50YdZ2UwVSS1FdL4j7TJEvRC6RHVdVxTfE1SVavqjnNMO61TfmVwNrDPr4WcNVo9TvMK0nqq67nTNNLQb8BXFhVhww7dQKwV/N6L+B/h5Xv2azq3Qq4Zf5w8MKYmUqSBt1TgT2APyU5tyl7N/Bx4JgkrwH+AbysOfcz4HnAJcCdwN5jNWAwlST1VdebNlTVKYw8DwojPLu7qgrY94G0YTCVJPWVG91LktTSIGx0bzCVJPXVIARTV/NKktSSmakkqa8G4akxBlNJUl+5AEmSpJYGYc7UYCpJ6qtBGOZ1AZIkSS2ZmUqS+mpoAHJTg6kkqa+cM5UkqaXJn5c6ZypJUmtmppKkvnKYV5Kklty0QZKkllzNK0lSS5M/lLoASZKk1sxMJUl95QIkSZJacs5UkqSWJn8oNZhKkvpsEIZ5XYAkSVJLZqaSpL5yzlSSpJYmfyg1mEqS+sw5U0mSZGYqSeqvGoCBXoOpJKmvBmGY12AqSeorV/NKktTS5A+lLkCSJKk1M9MB9rXDPsPzn7c91153PZts+iwAPvGxg3j+Ts/m7rvvZvbsy3nNa/fjlltu5ZGPXIs/n/9b/nbRbADOOGMW+77pwH52X7rXSz45k8c9c1Nuv+FWPv+cdwKw+hMeyS4f2YcZyyzF0Nwh/ve9h3PleZcCsN5Wj2en9+3B9BkzuOOm2/jaKz7Uz+5rDIMwzGtmOsCOOOIYnr/Tbvcr+9XJv2fjTZ7JZps/m4svns2B73zTvecunX05Wzx5B7Z48g4GUk0o5/zw9xy+1yfuV7bjga/k5M8fxxee925+dcgP2fFdrwRg2ZWWZ+cP7c0Rr/0Mn9vhAL77xs/3o8t6AIZaHhOBwXSA/eGUM7jxppvvV/bLX/2eefPmAXD6GbNYc83V+9E16QH5+5l/5c5bbr9fWQHLrLAcAMuutBy3XnMTAJu8cGsuOPEsbrnqBgDuuOHWJdpXPXDV8r+JoNNh3iQ7AT+rqonyx4OG2fvVu3LMD0649/16667DWWeexG233sb73v9JTjn1zD72ThrdTw4+gn2OOJDnvXs3Mi189SUfAGCVR63OtBnTed33D2KZBy3HqYefyB+P+0N/O6tRDUKA6HrOdFfg80mOBQ6vqgtHuzjJTGAmQKY/mGnTHtRx96audx34FubOnct3v3scAHPmXMt6j96SG2+8ic02fRLH/vCbbLTJM7jtttvHqEnqj612356ffOhILjjxLJ70/Kfwkk/M5Bu7f5Rp06ez5pPW4+uv+ihLLbs0bzjuYK7448Vcf9nV/e6yBlinw7xVtTuwKXApcHiS05LMTLLiQq4/rKq2qKotDKTd2WOPl/H8523PHnveN1969913c+ONvWGyWX/8E7Nn/50N1n9Uv7oojWmzlzydC048C4A//fQM1tq49/N6y9U3cNHvzueef93FnTfdxmVnXsgjHv/IfnZVYxiEYd7O50yr6lbgWOD7wOrAi4BZSd7cddv6T8/ZYTvesf8b2eXFr+Zf//r3veWrrPJQpk3r/Tist946POYx6zH7sn/0q5vSmG699ibW2+rxADx66ydyw9+vAeAvvziHdZ/8WKZNn8ZSyy7N2ps8husu+Wc/u6oxDMICpK7nTF8A7AM8GjgS2LKqrk2yPHAh8IUu25/qvnPkl9j26f/FKqs8lL/PPpuDP/hp3nnAm1hmmWU48effB+67BWabbbbiA+/fn7lz5zFv3jz2fdO7uGmBxUtSv+x66JtYb6vH86CHrMiBp32BX332WI478Ou84P17Mm3GNObedQ/HvevrAFx36VVc9LvzecuJH6eGirOP/g3XXHRln78CjWaoJkZ22Uaqwy8iyRHA16vq9yOce1ZVnbywz85Yes3J/93VlPeONbbtdxekxeJjf/9uuqp7j0e+uNXv+yMvP66zvo1Xp5lpVe05yrmFBlJJ0tQxCJlTp3OmSV6c5OIktyS5NcltSbzpS5J0ryGq1TERdH1rzCeBF4x1S4wkaeqaKCty2+g6mF5jIJUkjWairMhto+tgenaSo4EfAXfNL6yq4zpuV5KkJabrYLoScCeww7CyAgymkiRgMJ4a0/Vq3r27rF+SNPkNwpxp16t510pyfJJrk1yT5Ngka3XZpiRpchmEHZC63k7wcOAEYA1gTeDHTZkkSQBUVatjIug6mK5aVYdX1dzm+BawasdtSpK0RHUdTK9PsnuS6c2xO3BDx21KkiaRQdi0oetgug/wcuBqYA7w0qZMkiRgMOZMO1vNm2Q68JKqemFXbUiSJj9X846iquYBO3dVvyRJE0XXmzacmuSLwNHAHfMLq2pWx+1KkiaJiTLv2UbXwXTr5t8PDisr4JkdtytJmiQmyu0tbXS9A9IzuqxfkjT5db2IKMk3gZ2Aa6tqw2HlbwbeBMwFflpVBzTl7wJeA8wD3lJVJ43VRifBNMl+o52vqkO6aFeSNPksgQVI3wK+CBwxvyDJM+it69moqu5KslpT/gRgV+CJ9DYc+lWSDZp1QAvVVWa6Ykf1SpL0gFTV75Osu0DxG4CPV9VdzTXXNuU7A99vyi9LcgmwJXDaaG10Ekyr6uAu6pUkDZ62C5CSzARmDis6rKoOG+NjGwDbJPkI8G9g/6o6i97Wt6cPu+7KpmxUXS9AuleSWVW12ZJqT5I0ObRdgNQEzrGC54JmAA8BtgKeDByT5FFARmpiPJUtKSN1UJI0xfXp1pgrgeOqF8nPTDIErNKUrz3surWAq8aqrOvtBIf76RJsS5Kk0fyI5jbNJBsASwPX03vS2a5JlkmyHrA+cOZYlS2xzLSqDlpSbUmSJo+uV/Mm+R6wHbBKkiuB9wPfBL6Z5M/A3cBeTZZ6QZJjgL/Qu2Vm37FW8kLHwTTJi4FPAKvRG+YNUFW1UpftSpImj6GON22oqlcu5NTuC7n+I8BHHkgbXWemnwReUFUXdtyOJGmSmvz7H3UfTK8xkEqSRuPevGM7O8nR9CZ675pfWFXHddyuJElLTNfBdCXgTmCHYWUFGEwlSYDMjEEKAAAInUlEQVSZ6Ziqau8u65ckTX6D8NSYTu8zTbJWkuOTXJvkmiTHJlmryzYlSZPLENXqmAi63rThcHo3wK5Bb2/DHzdlkiQBvftM2/w3EXQdTFetqsOram5zfAtYteM2JUlaoroOptcn2T3J9ObYHbih4zYlSZNIVbU6JoKug+k+wMuBq4E5wEubMkmSgMGYM+16Ne8/gBd22YYkaXKbKNllG50E0yTvG+V0VdWHumhXkqR+6CozvWOEsgcBrwEeBhhMJUmAmzYsVFV9Zv7rJCsCbwX2Br4PfGZhn5MkTT0T5faWNjqbM03yUGA/YDfg28BmVXVTV+1Jkianrh/BtiR0NWf6KeDFwGHAk6rq9i7akSRNfoOQmXZ1a8zb6e16dBBwVZJbm+O2JLd21KYkSX3R1Zxp1/evSpIGhMO8kiS1NAjDvAZTSVJfmZlKktTSIGSmzm1KktSSmakkqa8c5pUkqaVBGOY1mEqS+qpqqN9daM05U0mSWjIzlST1lU+NkSSpJR8OLklSS2amkiS1NAiZqQuQJElqycxUktRXbtogSVJLbtogSVJLgzBnajCVJPXVIKzmdQGSJEktmZlKkvrKYV5JklpyNa8kSS0NQmbqnKkkSS2ZmUqS+moQVvMaTCVJfTUIw7wGU0lSX7kASZKklgZhO0EXIEmS1JKZqSSprxzmlSSpJRcgSZLU0iDMmRpMJUl9NQiZqQuQJElqycxUktRXg5CZGkwlSX01+UMpZBD+ItCiSTKzqg7rdz+ktvxZVr85Zzq1zex3B6TFxJ9l9ZXBVJKklgymkiS1ZDCd2pxj0qDwZ1l95QIkSZJaMjOVJKklg+mASXJ7v/sgjSRJJfnMsPf7J/nAYqr7A0n2Xxx1SYvCYDoFJJne7z5IwF3Ai5OssqQaTOLGNFoiDKYDKsl2SX6T5LvAn5qy/ZL8uTne1pStm+TCJF9LckGSXyRZLsmjk8waVt/6Sc7p05ejwTCX3kKh/1nwRJJHJjk5yfnNv+s05d9KcmiS/0syO8lLx2okyW+TfDTJ74C3PtC6kxyZZOdh9R2V5IWL6XugAWUwHWxbAu+pqick2RzYG3gKsBXwuiSbNtetD3ypqp4I3Ay8pKouBW5Jsklzzd7At5Zo7zWIvgTsluTBC5R/ETiiqjYCjgIOHXZudeBpwE7Ax8fZzspVtW1VfWYR6v46vZ93mn5uDfxsnO1qijKYDrYzq+qy5vXTgOOr6o6quh04DtimOXdZVZ3bvD4HWLd5/XVg72aY+BXAd5dMtzWoqupW4AjgLQuc+i/u+/k6kt7P63w/qqqhqvoL8PBxNnX0otZdVb8DHpNkNeCVwLFVNXec7WqKMpgOtjuGvc4o19017PU87nsAwrHAjvT+aj+nqm5YvN3TFPU54DXAg0a5Zvg9e8N/PgOQ5CNJzk1yLiO7YyHlY9bdOBLYjV6GevgodUmAwXQq+T2wS5LlkzwIeBHwh9E+UFX/Bk4CvoK/ULSYVNWNwDH0Aup8/wfs2rzeDThljDreU1WbVNUmo123KHU3vgW8rWnrgnFcrynOYDpFVNUser8gzgTOAL5eVX8cx0ePoveX/C+6652moM8Aw1f1voXelML5wB7AWxdjWw+47qq6BrgQ/4jUOLkDkkbV3Lv34Kp6b7/7Ii0pSZantwp+s6q6pd/90cTnPVhaqCTHA48GntnvvkhLSpLtgW8ChxhINV5mppIkteScqSRJLRlMJUlqyWAqSVJLBlMNhCTzmpv4/5zkB81qzEWta7skP2levzDJgaNcu3KSNy5CGyM+5WQ8Tz9p9pQdc4/aYdevm+TPD7SPksbPYKpB8a/mJv4NgbuB1w8/mZ4H/PNeVSdU1Wj7wa4MPOBgKmmwGEw1iP5Ab2/V+U/E+TIwC1g7yQ5JTksyq8lgVwBI8twkf01yCvDi+RUleXWSLzavH57k+CTnNcfW9DZHf3STFX+que4dSc5qnlJy8LC63pPkb0l+BTx2rC8iyeuaes5LcuwC2fb2Sf6Q5KIkOzXXT0/yqWFt//cIdT4xyZlNf89Psv4D//ZKWpDBVAMlvedX7kjz2Dl6QeuIqtqU3n6tBwHbV9VmwNnAfkmWBb4GvIDe5v+PWEj1hwK/q6qNgc2AC4ADgUubrPgdSXag9xSeLYFNgM2TPL15as+uwKb0gvWTx/HlHFdVT27au5D7b7+3LrAt8Hzgq83X8Brglqp6clP/65Kst0Cdrwc+32zDtwVw5Tj6IWkMbtqgQbHcsE3P/wB8A1gDuLyqTm/KtwKeAJyaBGBp4DTgcfSenHMxQJLvADNHaOOZwJ4AVTWP3iPqHrLANTs0x/ytGlegF1xXpPfUnjubNk4Yx9e0YZIP0xtKXoHePsnzHVNVQ8DFSWY3X8MOwEbD5lMf3LR90bDPnQa8J8la9IL1xePoh6QxGEw1KP614KbnTcBc8Mk5v6yqVy5w3Sbc/0kibQT4WFX9vwXaeNsitPEtYJeqOi/Jq4Hthp1bsK5q2n5zVQ0PuiRZ996Lqr6b5Ax6Ge1JSV5bVb9+gP2StACHeTWVnA48NcljoLf/apINgL8C6yV5dHPdKxfy+ZOBNzSfnZ5kJeA2elnnfCcB+wybi12zeS7m74EXJVkuyYr0hpTHsiIwJ8lS9J52MtzLkkxr+vwo4G9N229orifJBs0Tgu6V5FHA7Ko6FDgB2Ggc/ZA0BjNTTRlVdV2T4X0vyTJN8UFVdVGSmcBPk1xP7xFdG45QxVuBw5K8ht5zX99QVaclObW59eTnzbzp44HTmsz4dmD3qpqV5GjgXOByxnj8XeO99J7wczm9OeDhQftvwO/oPdD69VX17yRfpzeXOiu9xq8DdlmgzlcAuye5B7ga+OA4+iFpDO7NK0lSSw7zSpLUksFUkqSWDKaSJLVkMJUkqSWDqSRJLRlMJUlqyWAqSVJLBlNJklr6/zpLW7BLgvGMAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "from sklearn.metrics import confusion_matrix\n",
    "\n",
    "rf_cm = confusion_matrix(y_test, y_pred)\n",
    "# rf_cm\n",
    "\n",
    "def plotCM(cm):\n",
    "    ax = plt.subplot()\n",
    "    sns.heatmap(cm, annot=True, ax = ax, fmt=\"d\"); #annot=True to annotate cells\n",
    "\n",
    "    # labels, title and ticks\n",
    "    ax.set_xlabel('Predicted labels')\n",
    "    ax.set_ylabel('True labels')\n",
    "    ax.set_title('Confusion Matrix')\n",
    "    ax.xaxis.set_ticklabels(['Irony', 'Non-Irony']); ax.yaxis.set_ticklabels(['Irony', 'Non-Irony']);\n",
    "plotCM(rf_cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [],
   "source": [
    "# rf.get_params()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Gaussian NB"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 225,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.5688775510204082\n"
     ]
    }
   ],
   "source": [
    "from sklearn.naive_bayes import GaussianNB\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn import metrics\n",
    "\n",
    "gnb = GaussianNB()\n",
    "score = []\n",
    "# kfold = KFold(n_splits=10, random_state=1, shuffle=True)\n",
    "# for train, test in kfold.split(lexical_df):\n",
    "#     train_tf_idf = tweet_vectorizer.fit_transform(clean_data_df['clean_tweet'][train])\n",
    "gnb.fit(lexical_df, y_train)\n",
    "y_pred = gnb.predict(test_lex_df)\n",
    "print(metrics.accuracy_score(y_test, y_pred))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## SVM as it was in the paper"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn import svm \n",
    "from sklearn.model_selection import GridSearchCV\n",
    "\n",
    "def svc_param_selection(X, y, nfolds):\n",
    "    Cs = [0.001, 0.01, 0.1, 1, 10]\n",
    "    gammas = [0.001, 0.01, 0.1, 1]\n",
    "    param_grid = {'C': Cs, 'gamma' : gammas}\n",
    "    grid_search = GridSearchCV(svm.SVC(kernel='rbf'), param_grid, cv=nfolds)\n",
    "    grid_search.fit(X, y)\n",
    "    grid_search.best_params_\n",
    "    return grid_search.best_params_"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/plain": [
       "{'C': 1, 'gamma': 0.1}"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "svc_param_selection(lexical_df, y_train, 5)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {
    "scrolled": false
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "0.6683673469387755\n",
      "              precision    recall  f1-score   support\n",
      "\n",
      "           0     0.8000    0.6004    0.6860       473\n",
      "           1     0.5594    0.7717    0.6486       311\n",
      "\n",
      "    accuracy                         0.6684       784\n",
      "   macro avg     0.6797    0.6861    0.6673       784\n",
      "weighted avg     0.7046    0.6684    0.6712       784\n",
      "\n"
     ]
    }
   ],
   "source": [
    "\n",
    "clf = svm.SVC(C=1, gamma = 0.1, kernel = 'rbf')\n",
    "clf.fit(lexical_df, y_train)\n",
    "y_pred = clf.predict(test_lex_df)\n",
    "print(metrics.accuracy_score(y_test, y_pred))\n",
    "print(metrics.classification_report(y_test, y_pred, digits = 4))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x432 with 2 Axes>"
      ]
     },
     "metadata": {
      "image/png": {
       "height": 386,
       "width": 466
      },
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "%config InlineBackend.figure_format ='retina'\n",
    "svm_cm = confusion_matrix(y_test, y_pred)\n",
    "plotCM(svm_cm)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}