Data_Engineering.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "# This is the data engineering scripts for team 8\n",
    "## Group Members: Pryor Vo, Xiangyuan Ren, Peng Chen, Ge Chang, Siqi Qin"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "import warnings\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import ipywidgets as widgets\n",
    "# Draw inline\n",
    "%matplotlib inline\n",
    "sns.set_style(\"white\", {'ytick.major.size': 10.0})\n",
    "sns.set_context(\"poster\", font_scale = 1.1)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "warnings.filterwarnings('ignore')\n",
    "\n",
    "sessions= pd.read_csv('input/sessions.csv')\n",
    "train_users = pd.read_csv('input/train_users_2.csv')\n",
    "test_users = pd.read_csv('input/test_users.csv')\n",
    "age_gender_buckets=pd.read_csv('input/age_gender_bkts.csv')\n",
    "countries=pd.read_csv('input/countries.csv')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Basic Visualization and Pre-processing"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>id</th>\n",
       "      <th>date_account_created</th>\n",
       "      <th>timestamp_first_active</th>\n",
       "      <th>date_first_booking</th>\n",
       "      <th>gender</th>\n",
       "      <th>age</th>\n",
       "      <th>signup_method</th>\n",
       "      <th>signup_flow</th>\n",
       "      <th>language</th>\n",
       "      <th>affiliate_channel</th>\n",
       "      <th>affiliate_provider</th>\n",
       "      <th>first_affiliate_tracked</th>\n",
       "      <th>signup_app</th>\n",
       "      <th>first_device_type</th>\n",
       "      <th>first_browser</th>\n",
       "      <th>country_destination</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gxn3p5htnn</td>\n",
       "      <td>2010-06-28</td>\n",
       "      <td>20090319043255</td>\n",
       "      <td>NaN</td>\n",
       "      <td>-unknown-</td>\n",
       "      <td>NaN</td>\n",
       "      <td>facebook</td>\n",
       "      <td>0</td>\n",
       "      <td>en</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "      <td>Web</td>\n",
       "      <td>Mac Desktop</td>\n",
       "      <td>Chrome</td>\n",
       "      <td>NDF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>820tgsjxq7</td>\n",
       "      <td>2011-05-25</td>\n",
       "      <td>20090523174809</td>\n",
       "      <td>NaN</td>\n",
       "      <td>MALE</td>\n",
       "      <td>38.0</td>\n",
       "      <td>facebook</td>\n",
       "      <td>0</td>\n",
       "      <td>en</td>\n",
       "      <td>seo</td>\n",
       "      <td>google</td>\n",
       "      <td>untracked</td>\n",
       "      <td>Web</td>\n",
       "      <td>Mac Desktop</td>\n",
       "      <td>Chrome</td>\n",
       "      <td>NDF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4ft3gnwmtx</td>\n",
       "      <td>2010-09-28</td>\n",
       "      <td>20090609231247</td>\n",
       "      <td>2010-08-02</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>56.0</td>\n",
       "      <td>basic</td>\n",
       "      <td>3</td>\n",
       "      <td>en</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "      <td>Web</td>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>IE</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bjjt8pjhuk</td>\n",
       "      <td>2011-12-05</td>\n",
       "      <td>20091031060129</td>\n",
       "      <td>2012-09-08</td>\n",
       "      <td>FEMALE</td>\n",
       "      <td>42.0</td>\n",
       "      <td>facebook</td>\n",
       "      <td>0</td>\n",
       "      <td>en</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "      <td>Web</td>\n",
       "      <td>Mac Desktop</td>\n",
       "      <td>Firefox</td>\n",
       "      <td>other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>87mebub9p4</td>\n",
       "      <td>2010-09-14</td>\n",
       "      <td>20091208061105</td>\n",
       "      <td>2010-02-18</td>\n",
       "      <td>-unknown-</td>\n",
       "      <td>41.0</td>\n",
       "      <td>basic</td>\n",
       "      <td>0</td>\n",
       "      <td>en</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "      <td>Web</td>\n",
       "      <td>Mac Desktop</td>\n",
       "      <td>Chrome</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           id date_account_created  timestamp_first_active date_first_booking  \\\n",
       "0  gxn3p5htnn           2010-06-28          20090319043255                NaN   \n",
       "1  820tgsjxq7           2011-05-25          20090523174809                NaN   \n",
       "2  4ft3gnwmtx           2010-09-28          20090609231247         2010-08-02   \n",
       "3  bjjt8pjhuk           2011-12-05          20091031060129         2012-09-08   \n",
       "4  87mebub9p4           2010-09-14          20091208061105         2010-02-18   \n",
       "\n",
       "      gender   age signup_method  signup_flow language affiliate_channel  \\\n",
       "0  -unknown-   NaN      facebook            0       en            direct   \n",
       "1       MALE  38.0      facebook            0       en               seo   \n",
       "2     FEMALE  56.0         basic            3       en            direct   \n",
       "3     FEMALE  42.0      facebook            0       en            direct   \n",
       "4  -unknown-  41.0         basic            0       en            direct   \n",
       "\n",
       "  affiliate_provider first_affiliate_tracked signup_app first_device_type  \\\n",
       "0             direct               untracked        Web       Mac Desktop   \n",
       "1             google               untracked        Web       Mac Desktop   \n",
       "2             direct               untracked        Web   Windows Desktop   \n",
       "3             direct               untracked        Web       Mac Desktop   \n",
       "4             direct               untracked        Web       Mac Desktop   \n",
       "\n",
       "  first_browser country_destination  \n",
       "0        Chrome                 NDF  \n",
       "1        Chrome                 NDF  \n",
       "2            IE                  US  \n",
       "3       Firefox               other  \n",
       "4        Chrome                  US  "
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "train_users.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualization of Population in Destination Countries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "buckets=['0-4','5-9','10-14','15-19','20-24','25-29','30-34','35-39','40-44','45-49','50-54','55-59','60-64',  \\\n",
    "         '65-69','70-74','75-79','80-84','85-89','90-94','95-99','100+']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "73b56ddc77f94882bb5e32f7ac2ffe51",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.country_info>"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "gender=widgets.RadioButtons(\n",
    "    options=['male', 'female'],\n",
    "    description='Gender:',\n",
    "    disabled=False\n",
    ")\n",
    "destination=widgets.RadioButtons(\n",
    "    options=list(countries.country_destination),\n",
    "    description='Country:',\n",
    "    disabled=False\n",
    ")\n",
    "def country_info(gender, destination):\n",
    "    '''\n",
    "    This function performs the visualization of population in a certain country\n",
    "    You can switch among different destination countries and genders\n",
    "    *param gender: the gender of population you want to visualize\n",
    "    *param destination: the destination country of population you want to visualize\n",
    "    '''\n",
    "    country_pop=age_gender_buckets.loc[(age_gender_buckets['country_destination'] == destination) & (age_gender_buckets['gender']==gender),('age_bucket','population_in_thousands')]\n",
    "    pop=[float(country_pop.loc[country_pop['age_bucket']==i,'population_in_thousands']) for i in buckets]\n",
    "    y_pos = range(0,101,5)\n",
    "    y_val=pop\n",
    "    if gender=='female':\n",
    "        plt.bar(y_pos, y_val, align='edge', alpha=0.5,color='#F35FCD',width=4)\n",
    "    elif gender=='male':\n",
    "        plt.bar(y_pos, y_val, align='edge', alpha=0.5,color='#00AFFF',width=4)\n",
    "    plt.xticks(range(0,101,5), rotation = 70)\n",
    "    plt.xlabel('Ages')\n",
    "    plt.ylabel('Population in Thousands')\n",
    "    plt.show()\n",
    "widgets.interact(country_info,gender=gender,destination=destination)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visualization of Gender and Destination Related Stuffs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "count    250922.000000\n",
       "mean         49.668335\n",
       "std         155.666302\n",
       "min           1.000000\n",
       "25%          28.000000\n",
       "50%          34.000000\n",
       "75%          43.000000\n",
       "max        2014.000000\n",
       "Name: age, dtype: float64"
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "users = pd.concat((train_users, test_users), axis=0, ignore_index=True)\n",
    "users.gender.replace('-unknown-', np.nan, inplace=True)\n",
    "users.age.describe()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Distribution of destination country regarding gender"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaAAAAEiCAYAAAC/TgaKAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10d61dbd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "women = sum(users['gender'] == 'FEMALE')\n",
    "men = sum(users['gender'] == 'MALE')\n",
    "\n",
    "female_destinations = users.loc[users['gender'] == 'FEMALE', 'country_destination'].value_counts() / women * 100\n",
    "male_destinations = users.loc[users['gender'] == 'MALE', 'country_destination'].value_counts() / men * 100\n",
    "\n",
    "male_destinations.plot(kind='bar', width=0.4, color='#00AFFF', position=0, label='Male', rot=0)\n",
    "female_destinations.plot(kind='bar', width=0.4, color='#F35FCD', position=1, label='Female', rot=0)\n",
    "\n",
    "plt.legend()\n",
    "plt.xlabel('Destination Country')\n",
    "plt.ylabel('Percentage')\n",
    "\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Distribution of ages within 15 to 96"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAZgAAAEiCAYAAADEasRGAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nIiLiFgoYERFxCwWMiIi4hQJGRETc4v8DkIy8U1Lbb8wAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a139f4d10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "users.loc[users.age > 96, 'age'] = np.nan\n",
    "users.loc[users.age < 15, 'age'] = np.nan\n",
    "sns.distplot(users.age.dropna(), color='#FD5C64')\n",
    "\n",
    "plt.xlabel('Age')\n",
    "sns.despine()\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Country destination booking ratio of [destination] vs [All other countries excluding Not Defined]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "bd019f1cb4c647cf9157fcee4f41e6cf",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.find_destination_ratio>"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "x=widgets.RadioButtons(\n",
    "    options=list(countries.country_destination)+['other'],\n",
    "    value='US',\n",
    "    description='Country info from...',\n",
    "    disabled=False,\n",
    ")\n",
    "y=widgets.SelectionSlider(\n",
    "    options=[2,4,5,10,16,20],\n",
    "    value=5,\n",
    "    description='Age jumps (end point uninclusive)',\n",
    "    disabled=False,\n",
    "    continuous_update=False,\n",
    "    orientation='horizontal',\n",
    "    readout=True\n",
    ")\n",
    "def find_destination_ratio(destination,age_interval):\n",
    "    '''\n",
    "    This function shows the ratio of booking in given destination vs other countries (excluding NDF)\n",
    "    *param destination: the target country destination that you want to analyse with \n",
    "    *param age_interval: the age step for the x axis (age)\n",
    "    '''\n",
    "    #Establish different dataframes to compare\n",
    "    df=users.loc[users['country_destination'] == destination, 'age'];\n",
    "    df_comparison=users.loc[users['country_destination'] != destination, 'age']\n",
    "    df_comparison2=users.loc[users['country_destination'] == 'NDF', 'age']\n",
    "    \n",
    "    #Create age pools for better graph readability\n",
    "    bins=range(15,96,age_interval)\n",
    "    \n",
    "    #Value count for each pool in each dataframe, reindexed to be in order\n",
    "    count_interval=pd.cut(df,bins=bins,right=False).value_counts().reindex(pd.cut(df,bins=bins,right=False).cat.categories)\n",
    "    count_comparison=pd.cut(df_comparison,bins=bins,right=False).value_counts().reindex(pd.cut(df_comparison,bins=bins,right=False).cat.categories)-pd.cut(df_comparison2,bins=bins,right=False).value_counts().reindex(pd.cut(df_comparison2,bins=bins,right=False).cat.categories)\n",
    "    \n",
    "    #Put into numpy array to plot\n",
    "    users_booking_pools=np.array(count_interval,dtype=float)\n",
    "    users_booking_comparison=np.array(count_comparison,dtype=float)\n",
    "    users_booking_percent=users_booking_pools/users_booking_comparison*100\n",
    "    \n",
    "    #Plot\n",
    "    y_pos = xrange(15,95,age_interval)\n",
    "    y_val=users_booking_percent\n",
    "    plt.bar(y_pos, y_val, align='edge', alpha=0.5,color='#FFA35D',width=100/(len(xrange(15,96,age_interval))+5))\n",
    "    plt.xticks(xrange(15,96,age_interval), rotation = 90)\n",
    "    plt.xlabel('Ages')\n",
    "    plt.ylabel('Percent of Users Who Book for %s vs Other Countries' % x.options[x.index])\n",
    "    plt.show()\n",
    "widgets.interact(find_destination_ratio,destination=x,age_interval=y)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Comparison of different country bookings with a set age threshold"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "ee78c29431f84466a5e2c476be8e348c",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.find_destination_ratio_agethreshold>"
      ]
     },
     "execution_count": 10,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "\n",
    "age=widgets.SelectionSlider(\n",
    "    options=range(18,91,2),\n",
    "    value=50,\n",
    "    description='Age threshold at.....',\n",
    "    disabled=False,\n",
    "    continuous_update=True,\n",
    "    orientation='horizontal',\n",
    "    readout=True\n",
    ")\n",
    "\n",
    "def find_destination_ratio_agethreshold(age):\n",
    "    '''\n",
    "    This function shows the booking ratio of younger and older groups splitting with the given age threshold\n",
    "    *param age: the age threshold used to split younger and older.\n",
    "    '''\n",
    "    younger = sum(users.loc[users['age'] < age, 'country_destination'].value_counts())\n",
    "    older = sum(users.loc[users['age'] > age, 'country_destination'].value_counts())\n",
    "\n",
    "    younger_destinations = users.loc[users['age'] < age, 'country_destination'].value_counts() / younger * 100\n",
    "    older_destinations = users.loc[users['age'] > age, 'country_destination'].value_counts() / older * 100\n",
    "\n",
    "    younger_destinations.plot(kind='bar', width=0.4, color='#00FFFF', position=0, label='Youngers', rot=0)\n",
    "    older_destinations.plot(kind='bar', width=0.4, color='#FFA500', position=1, label='Olders', rot=0)\n",
    "\n",
    "    plt.legend()\n",
    "    plt.xlabel('Destination Country')\n",
    "    plt.ylabel('Percentage')\n",
    "\n",
    "    sns.despine()\n",
    "    plt.show()\n",
    "widgets.interact(find_destination_ratio_agethreshold,age=age)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this part, we shall specifically focus on the three items: affiliate_channel affiliate_provider first_affiliate_tracked\n",
    "The meaning of the three items are:\n",
    "    1. affiliate_channel: what kind of paid marketing\n",
    "    2. affiliate_provider: where the marketing is e.g. google, craigslist, other\n",
    "    3. first_affiliate_tracked: whats the first marketing the user interacted with before the signing up"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Here are all the potential country_destination:\n",
      "['NDF' 'US' 'other' 'FR' 'CA' 'GB' 'ES' 'IT' 'PT' 'NL' 'DE' 'AU']\n",
      "\n",
      "\n",
      "Here are all the potential affiliate_channel:\n",
      "['direct' 'seo' 'other' 'sem-non-brand' 'content' 'sem-brand' 'remarketing'\n",
      " 'api']\n",
      "\n",
      "\n",
      "Here are all the potential affiliate_provider:\n",
      "['direct' 'google' 'other' 'craigslist' 'facebook' 'vast' 'bing' 'meetup'\n",
      " 'facebook-open-graph' 'email-marketing' 'yahoo' 'padmapper' 'gsp' 'wayn'\n",
      " 'naver' 'baidu' 'yandex' 'daum']\n",
      "\n",
      "\n",
      "Here are all the potential first_affiliate_tracked:\n",
      "['untracked' 'omg' nan 'linked' 'tracked-other' 'product' 'marketing'\n",
      " 'local ops']\n",
      "\n",
      "\n"
     ]
    }
   ],
   "source": [
    "def showUnique(keyword):\n",
    "    '''\n",
    "    This function helps to show the potential unique choices \n",
    "    of a given keyword in the training dataset\n",
    "    * param keyword : the keyword of the information you want\n",
    "    '''\n",
    "    assert isinstance(keyword, str)\n",
    "    selected = train_users[keyword]\n",
    "    selected_unique = selected.unique()\n",
    "    print('Here are all the potential '+ keyword + ':')\n",
    "    print(selected_unique)\n",
    "    print('\\n')\n",
    "\n",
    "affiliate_keywords = ['country_destination', 'affiliate_channel', \\\n",
    "                  'affiliate_provider', 'first_affiliate_tracked']\n",
    "_ = map(showUnique, affiliate_keywords)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Visulization of Affiliate Related Stuffs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>country_destination</th>\n",
       "      <th>affiliate_channel</th>\n",
       "      <th>affiliate_provider</th>\n",
       "      <th>first_affiliate_tracked</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>NDF</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>NDF</td>\n",
       "      <td>seo</td>\n",
       "      <td>google</td>\n",
       "      <td>untracked</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>US</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>other</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>US</td>\n",
       "      <td>direct</td>\n",
       "      <td>direct</td>\n",
       "      <td>untracked</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "  country_destination affiliate_channel affiliate_provider  \\\n",
       "0                 NDF            direct             direct   \n",
       "1                 NDF               seo             google   \n",
       "2                  US            direct             direct   \n",
       "3               other            direct             direct   \n",
       "4                  US            direct             direct   \n",
       "\n",
       "  first_affiliate_tracked  \n",
       "0               untracked  \n",
       "1               untracked  \n",
       "2               untracked  \n",
       "3               untracked  \n",
       "4               untracked  "
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "utilize = train_users[['country_destination','affiliate_channel','affiliate_provider','first_affiliate_tracked']]\n",
    "utilize.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def uniqueCount_percentage(keyword, limit = -1):\n",
    "    '''\n",
    "    This is the function to visualize distribution of all unique items in train_user[keyword]\n",
    "    * param keyword : the keyword of the information you want\n",
    "    * limit : top limit counted will be considered, default value -1 means all should be counted\n",
    "    '''\n",
    "    print('Result for ' + keyword + '\\n')\n",
    "    all_keys = train_users[keyword]\n",
    "    all_counts = train_users[keyword].value_counts()\n",
    "    book_counts = train_users[train_users['country_destination'] != 'NDF'][keyword].value_counts()\n",
    "    if limit > 0:\n",
    "        all_counts = all_counts[ : limit]\n",
    "        book_counts = book_counts[ : limit]\n",
    "    \n",
    "    # deal with missing value in booking\n",
    "    for d in all_counts.index:\n",
    "        if d not in book_counts.index:\n",
    "            book_counts[d] = 0\n",
    "\n",
    "    per_counts = map(lambda x : book_counts[x] * 1.0 / all_counts[x], all_counts.index)\n",
    "    org_all = map(lambda x : all_counts[x], all_counts.index)\n",
    "    org_book = map(lambda x : book_counts[x], all_counts.index)\n",
    "    org_ndf = map(lambda x,  y: x - y, org_all, org_book)\n",
    "    org_dict = {'Book Customer': org_book, 'NDF Customer': org_ndf, 'Total Customer': org_all, 'Book Percent': per_counts}\n",
    "    df = pd.DataFrame(org_dict, columns = ['Book Customer', 'NDF Customer', 'Total Customer', 'Book Percent'], index = all_counts.index)\n",
    "    print(df)\n",
    "    ax = df[['Book Customer', 'NDF Customer']].plot(kind=\"barh\", stacked=True, figsize=(6, 8))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Result for affiliate_channel\n",
      "\n",
      "               Book Customer  NDF Customer  Total Customer  Book Percent\n",
      "direct                 59398         78329          137727      0.431273\n",
      "sem-brand              11094         14951           26045      0.425955\n",
      "sem-non-brand           7150         11694           18844      0.379431\n",
      "other                   3600          5361            8961      0.401741\n",
      "seo                     3955          4708            8663      0.456539\n",
      "api                     2785          5382            8167      0.341006\n",
      "content                  558          3390            3948      0.141337\n",
      "remarketing              368           728            1096      0.335766\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAekAAAHlCAYAAADRDATsAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQ5qIiEilGNJEREQq9X/RQk++OvlxpgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a1313bbd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "uniqueCount_percentage('affiliate_channel', -1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In this part, we can see that among all the affiliate channels, direct, seo are tht most two popular paid marketing ways. Paid search marketing means you advertise within the sponsored listings of a search engine or a partner site by paying either each time your ad is clicked (pay-per-click - PPC) or less commonly, when your ad is displayed (cost-per-impression - CPM). Therefore we can see that most customer come into airbnb without using mach paid marketing method."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Result for affiliate_provider\n",
      "\n",
      "                     Book Customer  NDF Customer  Total Customer  Book Percent\n",
      "direct                       59234         78192          137426      0.431025\n",
      "google                       20779         30914           51693      0.401969\n",
      "other                         4663          7886           12549      0.371583\n",
      "craigslist                    1616          1855            3471      0.465572\n",
      "bing                           833          1495            2328      0.357818\n",
      "facebook                       792          1481            2273      0.348438\n",
      "vast                           271           558             829      0.326900\n",
      "padmapper                      252           516             768      0.328125\n",
      "facebook-open-graph            141           404             545      0.258716\n",
      "yahoo                          174           322             496      0.350806\n",
      "gsp                             37           416             453      0.081678\n",
      "meetup                          46           301             347      0.132565\n",
      "email-marketing                 37           129             166      0.222892\n",
      "naver                           18            34              52      0.346154\n",
      "baidu                           10            19              29      0.344828\n",
      "yandex                           4            13              17      0.235294\n",
      "wayn                             0             8               8      0.000000\n",
      "daum                             1             0               1      1.000000\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAhkAAAHlCAYAAABYou50AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAGhpxb6O+vGOAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a130ece90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "uniqueCount_percentage('affiliate_provider', -1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this plot we can see that most paid marketings lie in the Google except the diret paid marketing way. It means we may invest more paid marketing ways which is highly related to the Google. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Result for first_affiliate_tracked\n",
      "\n",
      "               Book Customer  NDF Customer  Total Customer  Book Percent\n",
      "untracked              47739         61493          109232      0.437042\n",
      "linked                 20102         26185           46287      0.434290\n",
      "omg                    16425         27557           43982      0.373448\n",
      "tracked-other           2290          3866            6156      0.371995\n",
      "product                  542          1014            1556      0.348329\n",
      "marketing                 63            76             139      0.453237\n",
      "local ops                  9            25              34      0.264706\n"
     ]
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAdoAAAHlCAYAAABSyD3XAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAIABJREFUeJzs3XlYF1XD//E3oLiLG4qKmpR+NQHF\nMHNFIXO7NdJSyB3cwFyyO6XUXMFIcwlFRc19V3IJWxQX0kiTm1L7lZVFigvio7grovz+8GFuv+GK\nTKjP53VdXBec75kzZ0YvPpwzZ2ZsMjMzMxERERFT2OZ1B0RERJ5mCloRERETKWhFRERMpKAVEREx\nkYJWRETERApaEREREyloRURETKSgFRERMZGCVkRExEQKWhERERMpaEVEREykoBURETGRglZERMRE\nCloRERETKWhFRERMlC+vOyCPj6tXr3Lw4EEcHR2xs7PL6+6IiDwRbty4QWpqKq6urhQsWDDb5wpa\nMRw8eJAuXbrkdTdERJ5Iy5Ytw9PTM1u5glYMjo6OwK3/LE5OTnncGxGRJ8PJkyfp0qWL8Tv07xS0\nYsiaLnZycsLZ2TmPeyMi8mS52yU3LYYSERExkYJWRETERApaEREREyloRURETKSgFRERMZGCVkRE\nxEQKWhERERMpaEVEREykoBURETGRglZERMRECloRERETKWhFRERMpKAVERHTZGZm5nUX8pze3iMi\npnkmJCavu2Al6cO2Od62W7du7N2716qsYMGCVKxYkbZt29K3b1/y58//qF28o4iICGbMmMGuXbvu\n+iq2e/nrr79YtmwZcXFxnDx5kkKFCuHi4kKnTp1o164dtra5P+ZKSUkhNDSUwMBAateunevtP0kU\ntCIiD6hChQpMmjTJ+PnSpUv88MMPzJw5k5SUFMaNG5eHvbuzmJgYRowYQfny5fH398fFxYXLly+z\ndetWhg0bRlxcHOHh4eTLl7txsHv3br766isCAgJytd0nkYJWROQBFShQAE9PT6syLy8vTp06xdq1\naxk+fDhFihTJo95l9/PPPxMSEkK9evWYNWsWBQoUMD5r2bIlLi4uTJs2DU9PT/z9/fOwp083XaMV\nEXlEJUuWzFZ25coVZs6cSZs2bXB3d6dp06aMGzeOtLQ0q3ppaWlMnDiRV155BTc3N3x8fJg6dSrX\nrl276/7S09MJDAykTp067N69+6715syZw40bNxg/frxVyGYJDAzEz8+P0qVLA5CcnIzFYiEqKsqq\nXkxMDBaLhT179gC3rrtGRkbSqlUr3N3dqV+/PgMHDuTw4cPAranu9957D4DOnTvTrVs3o60ff/yR\n3r1706BBA2rXro2fnx/btm2z2p+3tzcTJ05k1qxZNGvWDDc3Nzp16sQvv/zCgQMH8Pf3x93dHR8f\nH1asWGG17eXLlwkPD6d58+a4urrSokUL4zxkiYiIwNPTkzVr1tCkSRM8PT3ZuXPnXc/jo9KIVkTk\nIWRkZBjfX716lYSEBNasWYOfn58xmk1PT6d79+78+uuv9O3bFzc3Nw4dOsSsWbOIj49nzZo1FC1a\nlLS0NN544w3OnTtHcHAwLi4uJCQkMH/+fBISEli4cGG2Kd309HSCgoJITExk3rx52UbYt9uxYwc1\na9akYsWKd/zc3t6esWPHPvQ5mDdvHjNnzmTQoEG4u7tz6tQppk+fTkBAAFu2bKFjx45kZGQwe/Zs\nxo0bR506dQDYsmULgwYNom7duowaNYp8+fKxatUqgoKCGDduHJ07dzb2sW7dOqpVq8bIkSO5ePEi\noaGhBAUFkZmZSUBAAL179+bTTz9l7NixuLm54erqyvXr1wkICODQoUP079+fGjVqkJCQwPTp00lK\nSmLixIlG+5cvXyYyMpIPPviAtLS0e57HR6WgFRF5QH/++Se1atXKVl6zZk3eeust4+cNGzawf/9+\nJk+eTLt27QBo2rQpNWrUoHfv3ixatIgBAwawYMECjhw5wvLly3nhhReMeuXLl2f06NF8/vnn+Pr6\nGu1mheyBAwdYuHAh7u7ud+3rmTNnuHLlCpUrV86twzfs2bMHZ2dn+vbti42NDQDOzs5s376dCxcu\nUKFCBapUqQKAxWLBYrGQmZnJhAkTqFq1KosWLTL+gGjRogX+/v589NFHtGvXjsKFCwNw48YNoqKi\nKFasGAAHDx5kyZIljB07Fj8/P2Of7du3JyEhAVdXVzZt2kRiYiLTpk2jdevWwK2p/VKlSjFx4kRj\nJJzV/uDBg2nRokWun5+/09SxiMgDqlixImvXrjW+li5dygcffMDp06d5/fXXSUlJAeDbb7/F3t6e\ntm2tVzk3adKE8uXLEx8fD0B8fDzPPPOMEbJZOnTogJ2dnVEvy5AhQ9i1axeBgYH3DFkAOzs7AKsp\n09zSpEkTkpKS8PX1ZcaMGSQkJFC7dm3+/e9/G9PQf/fHH39w8uRJ2rdvbzVKt7Gx4fXXX+fixYvs\n37/fKLdYLEbIApQtWxbA6lxlTdmfP38euLUAK1++fDRv3pyMjAzjq1WrVgDExcVZ9clisTzKaXhg\nGtGKiDwge3t73NzcrMrq1atH9erV6dq1K/Pnz+f9998nLS0NR0fHO9424+joaARDWloa5cqVu+N+\nSpQoYdTLcuLECerUqcPcuXN59dVXcXJyumtfHRwcKFasGEePHr3nMR0/fhxHR8eHujWpR48eODg4\nsG7dOmbPnk1ERAQlSpSgU6dODBkyxAj522Vdm84KzNtl3bJ04cIFo+xui8oKFSpkfJ81ms5y5swZ\nMjIy7no7UdYfQlnKlClzx3q5TUErIvKIsqaT//jjDwBKlCjBvn37uHnzZrawTUlJoWrVqka9U6dO\nZWsvPT2ds2fPZltklTWV2r59e0aMGMH8+fPv2S8vLy9iYmJITk7G2dk52+c3btygc+fOODg48Pnn\nnxvB9fdR8MWLF7Nt6+vri6+vL1euXGHfvn2sWbOGqKgonJyc6NKlS7b6JUqUALjj8WaV3WlR2cMo\nXrw4Dg4Odz0vj9p+Tj0xU8fe3t4EBgbm2f6jo6OxWCz88MMPedYHEXk8/ec//wEwArRBgwakp6cT\nE2P9wI5du3aRkpJC/fr1jXpJSUkkJCRY1YuOjubmzZtGvSyOjo5UqlTJmEJevnz5PfvVp08f7Ozs\nGD169B1XMc+aNYtTp07RqVMnAIoWLQrcGjnfLmu1cZYePXowYMAA4NYIs0mTJowfPx64tXIZyPYH\nhouLC05OTmzcuNFqQVlmZibr1q2jcOHCuLq63vN47qdhw4acO3cOGxsb3NzcjK+MjAw++ugjY1X0\nP00jWhGRB3Tt2jX27dtn/JyZmWmsJi5WrBg9evQAbo32Vq5cyciRIzly5Ahubm789ttvREZGUqVK\nFbp27QpAz5492bRpE8HBwQQFBeHi4kJiYiJz587Fw8ODNm3a3LEf3bp144svvmDSpEk0adKESpUq\n3bFejRo1GDlyJOPHj6djx474+fnh4uLCmTNn2Lx5M7Gxsfj6+hq33zg4OFCvXj3Wr19P9erVqVq1\nKlu2bMkWtI0aNeLjjz9mzJgxeHt7c/36dVasWEG+fPmMRUgODg7ArZXG+fPnp1atWoSEhPD222/T\no0cPunTpQv78+Vm1ahWJiYl88MEHFCxY8BH+dW6d91WrVtGvXz/69OlD9erVSUpKIiIigoIFC973\nurZZFLQiIg/o+PHjVtOi+fPnp1SpUrz00ksEBwcb07P29vYsXryYiIgI1qxZw6xZsyhTpgyvvfYa\nb731FsWLFwduTWWuWrWKadOmMW/ePNLS0qhQoQK9e/emf//+d71uamtrS2hoKK+99hohISEsWbLk\nro9R9Pf35/nnn2fp0qUsWLCA06dPU7RoUVxcXJg+fTotW7a0utYZHh5OWFgYH3/8Mba2tjRr1ozJ\nkyfTs2dPo07fvn0pVKgQq1evZsOGDQC4uroyf/58I8xeeuklmjZtyuLFi4mLi2PTpk20bt2a4sWL\nM3v2bN5//31sbGx4/vnniYyMxMfHJ+f/MP+rQIECxnlftGgRqamplCpVimbNmjFo0KA8mzq2yXxC\nnvjs7e1N1apVrebe4+LiiIqK4tdffyUjI4M6deowYMAAq1VpN2/eZNGiRaxbt44jR45QsmRJWrRo\nweDBg40VbUlJSURGRvLdd9/xP//zPxQqVIjatWszePBg4z9NdHQ07733HqtWrTLuCbuTkydP8skn\nn/Ddd9+RmpqKs7Mzr7/+Oj179jQWCISEhHDw4EHeeecdJk+ezNGjR3F2dqZ79+7GsnWAX3/9lfDw\ncH766ScuXbpElSpV6NixIz179sy2CCA3JCcn4+PjQ2xs7B2v54iISHb3+935xI5oFy1aRFhYGD4+\nPoSFhXH16lUWLFhA9+7diYyMxMvLC4BRo0axbt06unTpwrvvvktycjJTpkzh999/Z+HChZw5c4bO\nnTtToUIFhg8fTqlSpTh8+DCRkZEMGDCA2NhY7O3tH6hPR44coVOnThQsWJCgoCCcnJzYuXMnkyZN\n4v/9v//Hxx9/bNQ9fvw477zzDv3796datWp8/vnnjB49mvPnz9O3b18uXrxIr169qFSpEuPGjaNI\nkSJ8/fXXfPjhh9jb299xsYGIiDx+nsigvXDhAlOnTqVRo0ZERkYa5S1atKBNmzaMHz8eLy8v/vzz\nT9auXUuPHj14//33jXoFChRgzpw5HDt2jKSkJKpUqcKUKVOMv0QaNGjAtWvXjIvnNWvWfKB+ffLJ\nJ1y6dIk1a9YY10y8vLwoUqQIUVFR+Pv7G08fuXTpEpMmTaJ9+/YANG/enLS0NGbNmkX37t05fPgw\np0+fZvjw4bzyyivAresiJUuWzNH0R4cOHe5bJz09/aHbFRGRe3sigzYxMZErV67w2muvWZUXKFCA\n9u3bExkZydGjR41XWv39pvHXX3+d119/Hbh1A3qjRo3IzMzk+PHjHDlyhD///JMdO3YADxc+8fHx\n1KtXL9vChE6dOhEVFcW3335rBG3hwoWNJ8Zk+de//sWuXbvYv38/bm5ulCtXjlGjRrFr1y4aNmxI\nw4YNGTJkyAP3R0RE8t4TGbQPeuPz2bNngfvflDxnzhwWL17M6dOnKVGiBBaLxVj99jCXsNPS0u7Z\np9tvPi9Xrly266xZT1Q5d+4chQoVYtWqVcyZM4ft27cbCw48PDwYMWJEtpvm7yc6Ovq+dbKuM4iI\nSO55Yu6jvd29bnzOevJHyZIljcVO//M//2NV5/Lly+zcuZMzZ84wd+5cpkyZQu/evdm9ezd79uxh\n8eLFNGvWLEf9ul+fsmT9EXC71NRU4L+BW758ecaMGcPOnTv5+uuvGTFiBH/99Rf9+/fn5s2bD90/\nERH55z2RQevh4UGhQoX47LPPrMrT09PZtGkTlStXpnz58sbN3l988YVVva+//pq+ffuSlJTE999/\nT5kyZejVq5fVyDfrtU0PE2gvvfQS33//fbZHnq1duxbA6ubztLQ0vv32W6t6mzZtokSJEri5ubF9\n+3YaNGjAgQMHAKhSpQrdu3enbdu2nD59mitXrjxwv0REJO88kVPHxYoVY9CgQYSHhxMcHEyHDh24\ndu0aCxcu5MSJE8yYMQOA5557jo4dO7Jo0SLg1iKnI0eO8Mknn+Dl5YWHhwceHh7s3LmTsLAwmjVr\nRlpaGmvXrjVC8GECbdCgQcTFxdGtWzeCg4MpV64c33zzDcuWLaNly5bZXsM0dOhQBg4cSIUKFdiw\nYQPx8fGEhYWRP39+6tati52dHW+//TZBQUFUqFCBX3/9lejoaLy9vR+rl0uLiMjdPZFBCxAQEEC5\ncuVYsGABQ4cOxd7enjp16rBkyRKrQBs/fjzPPPMM69atY8mSJZQtW5bOnTsTFBSEjY0NvXv35sKF\nC8TExLBixQrKlCmDh4cHK1euxN/fn71799KoUaMH6lOVKlVYs2YN06ZN4+OPP+by5ctUqVKFd999\n13hiTBZ7e3vCwsIIDw/nxIkTVK9enZkzZ/Lyyy8Dt56qsnjxYqZPn87UqVON67+dO3dm4MCBuXci\nRUTEVE/MAyueJiEhIcTExBjTwo8LPbBCROTh3e935xN5jVZERORJoaAVEZFco0nS7J7Ya7RPsg8/\n/JAPP/wwr7shYr4xDnndA2tjzuV405CQED777DNGjRplvH3ndjExMQwdOpTFixcbdxh4e3tz7Ngx\no46NjQ0FCxakfPnyeHt706dPH+N2RXiwe9lnz55N8+bN71nnr7/+YtmyZcTFxXHy5EkKFSqEi4sL\nnTp1ol27dnd9AcGjSElJITQ0lMDAwLu+eP3/KgWtiMhDmDRpEg0aNODZZ599oPq1a9dm2LBhwK3R\n3sWLF0lMTGTRokV8+eWXLF++nHLlyllt4+fnl+3JcVmqVat2z/3FxMQwYsQIypcvj7+/Py4uLly+\nfJmtW7cybNgw4uLiCA8PJ1++3P31v3v3br766isCAgJytd2ngYJWROQB2dnZYWdnx7vvvsuqVavu\n+hq72xUrVizbrX3NmzenUaNGBAQEMHbsWKtntsOtR8P+fZsH8fPPPxMSEkK9evWYNWsWBQoUMD5r\n2bIlLi4uTJs2DU9PT/z9/R+6fckZXaMVEXlAdnZ2jBgxgp9++sm4Xz+n6tevj7e3N7GxsRw/fjxX\n+jdnzhxu3LjB+PHjrUI2S2BgIH5+fsbT55KTk7FYLERFRVnVi4mJwWKxGC98z8zMJDIyklatWuHu\n7k79+vUZOHAghw8fBiAiIoL33nsPgM6dOxsvkgf48ccf6d27Nw0aNKB27dr4+fkZDwTK4u3tzcSJ\nE5k1axbNmjXDzc2NTp068csvv3DgwAH8/f1xd3fHx8eHFStWWG17+fJlwsPDad68Oa6urrRo0cI4\nD1kiIiLw9PRkzZo1NGnSBE9PT3bu3JnT0/zQNKIVEXkIHTt2ZPv27cydOxcvLy/q1q2b47aaNm3K\n119/zd69e/H19TXKb968SUZGRrb6dnZ293wX9Y4dO6hZsyYVK1a84+f29vaMHTv2ofs5b948Zs6c\nyaBBg3B3d+fUqVNMnz6dgIAAtmzZQseOHcnIyGD27NmMGzfOeGf3li1bGDRoEHXr1mXUqFHky5eP\nVatWERQUxLhx4+jcubOxj3Xr1lGtWjVGjhzJxYsXCQ0NJSgoiMzMTAICAujduzeffvopY8eOxc3N\nDVdXV65fv05AQACHDh2if//+1KhRg4SEBKZPn05SUhITJ0402r98+TKRkZF88MEHpKWl5WjGIKcU\ntCIiD2ncuHH88MMPDBs2jA0bNuT4SW1ZLyHJes55lqlTpzJ16tRs9SdOnHjXV16eOXOGK1euULly\n5Rz15V727NmDs7Mzffv2NYLe2dmZ7du3c+HCBSpUqECVKlUAsFgsWCwWMjMzmTBhAlWrVmXRokXG\nNeEWLVrg7+/PRx99RLt27ShcuDAAN27cICoqynhG/cGDB1myZAljx47Fz8/P2Gf79u1JSEjA1dWV\nTZs2kZiYyLRp02jdujVw69WkpUqVYuLEicZIOKv9wYMH06JFi1w/P/ejqWMRkYdUqlQpQkNDOXr0\nKKGhoY/c3t9HqV27dmXt2rXZvu612tjOzg7Aaso0tzRp0oSkpCR8fX2ZMWMGCQkJ1K5dm3//+9/G\nNPTf/fHHH5w8eZL27dtbLbyysbHh9ddf5+LFi+zfv98ot1gsRsjCf/8IeeGFF4yyrBezZL0Jbffu\n3eTLl4/mzZuTkZFhfLVq1QqAuLg4qz5ZLJZHOQ05phGtiEgOeHl54efnx8qVK/H29s5RGydPngTA\nycnJqrxcuXIP/SpMBwcHihUrlu2lJn93/PhxHB0dH2ghV5YePXrg4ODAunXrmD17NhEREZQoUYJO\nnToxZMgQI+Rv96CvM81yt1mBQoUKGd///Q+SM2fOkJGRcdfbibLenJblfq9MNYuCVkQkh0JCQvju\nu+8YNWoU/fv3f+jtd+/ejZ2dHfXq1cuV/nh5eRETE0NycvIdHwV448YNOnfujIODA59//rkRXH8f\nBV+8eDHbtr6+vvj6+nLlyhX27dvHmjVriIqKwsnJiS5dumSrf6/XmWaV3f7q0JwoXrw4Dg4OzJ8/\n/46fP2r7uUVTxyIiOVSoUCEmTZrE+fPnH3oV8o8//khsbCwtW7bMdh9tTvXp0wc7OztGjx7NtWvX\nsn0+a9YsTp06RadOnQAoWrQoACdOnLCql7XaOEuPHj0YMGAAcOuYmzRpwvjx44FbK5eBbA/BcHFx\nwcnJiY0bN1ot7MrMzGTdunUULlwYV1fXRzlcGjZsyLlz57CxscHNzc34ysjI4KOPPjJWRec1jWhF\nRB6Bu7s7/fv3v2vQXrhwgX379gG3QubChQskJiaydOlSnJ2dGTlyZK71pUaNGowcOZLx48fTsWNH\n/Pz8cHFx4cyZM2zevJnY2Fh8fX2N228cHByoV68e69evp3r16lStWpUtW7ZkC9pGjRrx8ccfM2bM\nGLy9vbl+/TorVqwgX758xiIkB4dbTwHbsmUL+fPnp1atWoSEhPD222/To0cPunTpQv78+Vm1ahWJ\niYl88MEHFCxY8JGO19fXl1WrVtGvXz/69OlD9erVSUpKIiIigoIFCxoLofKaglZE5BEFBQURFxdn\ntbgny48//mg1tVq4cGEqV65MQEAAvXr1MkaVucXf35/nn3+epUuXsmDBAk6fPk3RokVxcXFh+vTp\ntGzZ0upaZ3h4OGFhYXz88cfY2trSrFkzJk+eTM+ePY06ffv2pVChQqxevZoNGzYA4Orqyvz5840w\ne+mll2jatCmLFy8mLi6OTZs20bp1a4oXL87s2bN5//33sbGx4fnnnycyMvK+j5p8EAUKFGDx4sVE\nRESwaNEiUlNTKVWqFM2aNWPQoEGPzdSxXpMnBr0mT0Tk4ek1eSIiInlIQSsiImIiBa2IiIiJFLQi\nIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS0IqI\niJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQioiImEhBKyIi\nYiIFrYiIiIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiI\niRS0IiIiJlLQioiImEhBm8syMzPzugsiIvIY+T8ftBEREVgsFlJTUx+5re3bt9O/f3/j5+TkZCwW\nC1FRUY/ctoiIPJny5XUHniaffvqpVWCXLVuWZcuW4ezsnIe9EhGRvKSgNZG9vT2enp553Q0REclD\nj9XUsbe3NxMnTmTWrFk0a9YMNzc3OnXqxC+//MKBAwfw9/fH3d0dHx8fVqxYYWyXlJTEsGHDaNq0\nKbVq1cLT05PAwED2799v1ImOjsZisbBx40ZatGhBnTp1WL16dbY+pKenExgYSJ06ddi9e7dR/t13\n39GtWzc8PDzw8PCgb9++HDp0yKrve/fu5c8//8RisRAdHZ1t6njPnj1YLBZ27tzJO++8Q7169ahT\npw69e/fm8OHDVv347bff6NevH56entSrV4/33nuPpUuXYrFYSE5OzrVzLiIi5nqsghZg3bp1xMXF\nMXLkSMaPH8+ff/5JUFAQAwcOpHXr1kydOhUnJyfGjh3LwYMHOXPmDJ07d+a3335j+PDhzJs3jyFD\nhvDzzz8zYMAA0tPTrdoPDQ1l0KBBTJgwgcaNG1t9lp6eTlBQEImJicybN49GjRoBEBsbS0BAADY2\nNoSHhzNu3DhSU1Px8/Pj119/BWD69OnUrFmTChUqsGzZMry8vO56jMOGDaNYsWJMnjyZ999/nx9+\n+IGBAwcaC6mOHTuGv78/SUlJjBo1igkTJvDXX38xadKk3DzVIiLyD3jspo5v3LhBVFQUxYoVA+Dg\nwYMsWbKEsWPH4ufnB4CzszPt27cnISGBc+fOUaVKFaZMmWJcC23QoAHXrl3jo48+4vDhw9SsWdNo\nv1u3brRr1y7bfrNC9sCBAyxcuBB3d3fg1iri0NBQnnvuOT799FPy5bt1yry9vWnVqhWTJ08mKioK\nNzc3ihUrxtWrV43p4ruNPL28vBgzZozx84kTJ4iMjOSPP/7g2WefZdasWVy/fp3FixdTrlw5AJo3\nb07btm05cuRIjs5rhw4d7lvn73+UiIjIo3vsRrQWi8UIWbi1oAjghRdeMMpKliwJwPnz52nUqBGr\nV6+mYsWKHD9+nO+++44VK1awY8cOIHt41KhR4477HTJkCLt27SIwMNAIWbg1LX3s2DF8fHwAyMjI\nICMjgwIFCtCsWTPi4+O5fv36Qx1jvXr1rH6uWLEiAJcuXQJg165dvPjii0bIwq3rvf/6178eaj8i\nIpL3HrsRbZEiRe5YXqhQIeN7Gxsbq8/mzJnD4sWLOX36NCVKlMBisVCwYEEg+32tZcqUuWP7J06c\noE6dOsydO5dXX30VJycnAM6cOQNAZGQkkZGRd9z2zJkzVqF4P4ULF7b62db21t87N2/eBOD06dM0\naNAg23ZZf3TkRHR09H3rJCcnG39QiIhI7njsgvZhzZ07lylTphASEkK7du2MIF22bBlxcXEP3E7W\ndHX79u0ZMWIE8+fPB8DBwQGAPn360LJlyztumzXCzi0ODg6cOnUqW/np06dzdT8iImK+x27q+GF9\n//33lClThl69elmNVrdt2wb8d5R4P46OjlSqVMmYQl6+fDkALi4ulCtXjt9//x03Nzerr5UrV7J4\n8WLy588P/Hdk+qgaN27Mvn37SElJMcpu3LjBF198kSvti4jIP+eJD1oPDw9Onz5NWFgY3377LZs3\nbyYgIMC4NefKlSsP1V7WLTyTJk3i6NGj2NraEhISwo4dOxg0aBBbtmxh586dDB48mLVr11KjRg1j\nKtvBwYGUlBS2b99uFZIP66233iJ//vz06NGDDRs2sHXrVvr06UNSUhKQe4EuIiLme+J/Y/fu3ZvA\nwEC++uor+vXrx6RJkyhRogQrV67E1taWvXv3PlR7tra2hIaGcuPGDUJCQrh58yZt2rRh/vz5nD17\nlmHDhjFkyBCSk5MJDw8nMDA/BYRJAAAgAElEQVTQ2LZbt26UKlWKgQMHsmHDhhwfU6VKlVi6dCkV\nKlRgzJgxhISEUL58efz9/YHs13hFROTxZZOpp+A/duLi4sjMzMx2L25wcDCJiYnEx8ebst+sxVCx\nsbF6bKSIyAO63+/OJ34x1NPol19+YcqUKfTr148XX3yRjIwMtm3bRmxsLMOHD8/r7omIyENQ0D6G\nAgMDyZ8/P+vXr2fx4sUAVKtWjY8++ohXX301j3snIiIPQ0H7GLKzs6NXr1706tUrr7siIiKP6Ilf\nDCUiIvI4U9CKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS\n0IqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQioiImEhB\nKyIiYiIFrYiIiIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWt\niIiIiRS0IiIiJlLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQi\nIiImypfXHZDHT+Pw7SSVfuu/BWPO5V1nRESecBrRioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS0IqI\niJhIQSsiImIiBa2IiIiJFLQiIiImUtA+QTIzM/O6CyIi8pAUtLkoIiICi8VCampqrre9fft2+vfv\nn+vtioiIufQIxifEp59+akqAi4iIuTSiFRERMdFTGbTe3t6MGzeOOXPm0KRJE+rUqcObb77Jnj17\nAEhOTsZisTB//nw6d+6Mu7s7ISEhAFy5coWZM2fSpk0b3N3dadq0KePGjSMtLc1qH3/99RcDBw6k\nXr161KtXjwkTJnD9+nWrOnebSg4MDMTb29uqbPv27XTp0gUPDw9eeuklBgwYwB9//GEcz969e/nz\nzz+xWCxER0fn6vkSERHzPLVTxzExMRQuXJihQ4dSqFAhoqKiCAgIYNGiRTg5OQEwdepUgoKCCAoK\nomjRoqSnp9O9e3d+/fVX+vbti5ubG4cOHWLWrFnEx8ezZs0aihYtypkzZ/Dz86NgwYK89957FCtW\njKVLl5KQkJCjvq5fv57hw4fTuHFjwsPDycjIICIigu7du7Nx40amT5/OqFGjOHfuHJMmTaJq1aoP\nvY8OHTrct056enpOui8iIvfw1AbtpUuXWLt2LZUqVQKgcePGvPzyy3zyySeEhYUBULduXQYMGGBs\ns2bNGvbv38/kyZNp164dAE2bNqVGjRr07t2bRYsWMWDAABYtWsTZs2fZuHEj1atXB6B58+a0bduW\npKSkh+pnZmYmU6ZM4fnnn2fevHnY2NgAUKNGDXr16sW+fft45ZVXKFasGFevXsXT0/NRT42IiPyD\nntqgbdiwoRGyAEWLFqV58+Zs3LiRmzdvAmCxWKy2+fbbb7G3t6dt27ZW5U2aNKF8+fLEx8czYMAA\n9uzZg4uLixGyAPny5aNNmzZERkY+VD+TkpJISUmha9euRsgCuLi4sHPnzodq614eZLo5OTkZHx+f\nXNuniIg8pddoAWN6+HalS5cmIyODixcvAuDo6Gj1eVpaGo6OjtjaZj8tjo6OnD9/HoCzZ89SunTp\nbHXKli370P08e/bsHfsiIiJPh6c2aLMC7HapqanY29tTrFixO25TokQJUlNTjRHv7VJSUihZsiQA\npUqVuuOtNmfOnLH6OWuE+vf2Ll26ZHxfvHhxAE6fPp2tvV27dj30VLSIiDxentqgjY+PtwrbCxcu\nsGPHDho1amQ1RXu7Bg0akJ6eTkxMjFX5rl27SElJoX79+sCt671//vkn+/fvN+pkZmYSGxtrtV3R\nokUBOH78uFGWlpbGL7/8Yvzs4uKCo6MjX331ldW2x44do0+fPnz55ZcAdxxli4jI4++p/e196dIl\nevTowebNm9m8eTPdu3cnPT2doUOH3nUbX19fatWqxciRI5k5cyZxcXHMnz+fwYMHU6VKFbp27QpA\njx49cHZ2JigoiJUrV7Jt2zaCg4ON23GyeHt7Y2try/jx44mNjeWLL74gICCAwoULG3VsbW3597//\nzYEDBwgKCmLr1q3ExMTQr18/nJyc6NSpEwAODg6kpKSwfft2UlJSTDhjIiJihqc2aBs2bEjr1q2Z\nMGECI0eOpGzZsqxcudJqAdPf2dvbs3jxYvz8/FizZg3BwcEsWbKE1157jdWrVxvTvEWLFmXFihU0\naNCAKVOm8M4771C4cGEGDhxo1V7lypWZOnUq169fZ/DgwUyePJn27dvz6quvWtXz9fVl5syZnD59\nmqFDhxIaGorFYmHJkiWUKlUKgG7dulGqVCkGDhzIhg0bcvlsiYiIWWwyn8In1Xt7e1O1alXmz5+f\n1115omStOr72ykiSSr/13w/GnMu7TomIPOayfnfGxsbi7Oyc7fOndkQrIiLyOFDQioiImOipfGDF\ntm3b8roLIiIigEa0IiIiplLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZ6Klcdy6PZNbw5OOshFSIi\nuUEjWhERERMpaEVEREykoBURETGRglZERMRECloRERETKWhFRERMpKAVERExkYJWRETERApaERER\nEyloRURETKSgFRERMZGCVkRExEQKWhERERMpaEVEREykoBURETGRglZERMRECloRERETKWhFRERM\npKAVERExkYJWRETERApaEREREyloRURETKSgFRERMZGCVkRExEQKWhERERMpaEVEREykoBURETGR\nglZERMRECloRERETKWhFRERMlC+vOyCPn8bh26FIqWzlSQXffPBGxpzLxR6JiDy5NKIVERExkYJW\nRETERApaEREREyloRURETKSgFRERMZGCVkRExEQKWhERERMpaEVEREz0QEGbmZlpdj+eeDpHIiJy\nJ/cN2sTERPz9/f+JvgAQERGBxWIhNTXVlPaTk5OxWCxERUXlSnvnz59n1KhRfPXVV0ZZSEgIbm5u\nudK+iIg82e4btKtWreKnn376J/ryRPr5559ZvXo1N27cyOuuiIjIY0jXaEVEREx0z6Dt1q0bn332\nGenp6VgsFmNad+rUqQQGBlK7dm26d+8OwKlTpxgzZgze3t64urri4eGBv78/cXFxVm2eP3+eCRMm\n0KxZM9zd3WndujWLFi26ax/S09MJDAykTp067N692yj/7rvv6NatGx4eHnh4eNC3b18OHTpktW1G\nRgYzZszA29sbd3d3/Pz8+OOPPx745KSlpTFx4kReeeUV3Nzc8PHxYerUqVy7dg2A6Oho4/iHDh2K\nt7e31fY7d+6kQ4cOuLm50aRJEyZNmkR6erpVna+++oo33ngDd3d3XnzxRd5++22OHTtmfJ411T1/\n/nw6d+6Mu7s7ISEhD3wMIiKSt+759p6RI0cyefJk4uPjWbhwIRUqVGDGjBnMnz+fTp060bNnT27e\nvEl6ejrdunUjIyODt956i/Lly3Ps2DHmzJnDwIED2bZtG6VLl+batWu8+eabnDhxgqCgIGrUqMG3\n335LWFgYly5dIjg42Gr/6enpBAUFkZiYyLx58/D09AQgNjaWgQMH4unpSXh4ONeuXePTTz/Fz8+P\nVatWUb16dQBGjRrFhg0bCAgIoF69eiQkJDBo0KAHOjFpaWm88cYbnDt3juDgYFxcXEhISGD+/Pkk\nJCSwcOFCvLy8GDlyJBMmTGDgwIF4eXkZ21+/fp0RI0YQHBxM5cqViYmJYd68eRQpUsQ4zqVLlzJ+\n/HhatmxJ//79SUtLY+bMmXTu3Jno6GjKli1rtDd16lSCgoIICgqiaNGiD3QMt+vQocN96/z9jwAR\nEXl09wxai8VC6dKlsbGxMUIOoHz58owcORJb21sD4l9++YWSJUsybNgw6tata9QrWrQogwcPJjEx\nkZdffpn169fz22+/MXPmTF5++WUAGjduzLlz54iPj7cK2qyQPXDgAAsXLsTd3R24tbo3NDSU5557\njk8//ZR8+W4dgre3N61atWLy5MlERUWRlJREdHQ0gYGB/Pvf/wbAy8sLe3t7IiIi7ntiFixYwJEj\nR1i+fDkvvPACAE2bNqV8+fKMHj2azz//HF9fXyPUq1atarUAKjMzk/DwcBo1agRAw4YN2b17N998\n8w3BwcFcvHiRKVOm0LRpUz755BNju8aNG9OyZUtmzZrF6NGjjfK6desyYMCA+/ZbREQeLzl6H221\natWMkAWoUaMGK1euBCAlJYWjR4/y119/sW3bNuC/I6W9e/dSsGBBI2SzhIaGZtvHkCFD2L9/P0OH\nDjVCFiApKYljx44ZoZyRkQFAgQIFaNasGevXr+f69evs2bMHgFatWlm1++qrr1oFbXp6erZbcwoU\nKEB8fDzPPPOMEbJZOnTowLhx44iPj8fX1/ee5+nFF180vre1taVSpUqcPXsWgB9++IFLly7RsmVL\n4xgASpcuzQsvvJBtyt1isdxzX/cTHR193zrJycn4+Pg80n5ERMRajoK2TJky2crWrFnDrFmzOHbs\nGEWLFqVatWqUK1cO+O89pmfPnqV06dIPtI8TJ05Qp04d5s6dy6uvvoqTkxMAZ86cASAyMpLIyMg7\nbnvmzBkj0P6+v9unYwECAwPZu3evVdmhQ4dIS0sz+n87e3t7SpQowfnz5+/Zf3t7e/Lnz29VZmtr\ny82bN62OY8SIEYwYMSLb9n/f1tHR8Z77ExGRx1OOgvbvNm/ezMiRIwkICKB79+6UL18egB07dvDl\nl18a9YoVK2YEzO2OHj3KsWPH8PDwMMqioqIoVqwY7du3Z8SIEcyfPx8ABwcHAPr06UPLli3v2J+S\nJUsaAZuamkrFihWNz/6+/7Fjx3Lp0qVsbZQoUYJTp05lK09PT+fs2bOULFnyzifjAWUdx4gRI6yO\nW0REni73vb3n9iniu/n+++8BGDx4sBGygDF1nDWKe/HFF7ly5Qo7duyw2n7WrFnZFkI5OjpSqVIl\nhgwZwq5du1i+fDkALi4ulCtXjt9//x03Nzerr5UrV7J48WLy589Pw4YNsbW1ZcOGDVbtfv3111Y/\nu7i4ZGsHoEGDBiQlJZGQkGBVPzo6mps3b1K/fn0A7Ozs7nt+7sTDw4OCBQvy119/We27Zs2azJ49\nm5iYmBy1KyIij5f7jmgdHBxIT0/n888/t7pWejsPDw+WL1/OyJEj8fX15erVq8TExPDFF18AcOXK\nFeDW9c2VK1cybNgwBgwYwLPPPst3331HdHQ077zzDgUKFMjWdrdu3fjiiy+YNGkSTZo0oVKlSoSE\nhDB06FAGDRpEu3btsLe3Jzo6mi+//JJhw4ZhY2NDxYoV6dmzJwsWLMDOzo4mTZrw008/3fNWotv1\n7NmTTZs2ERwcTFBQEC4uLiQmJjJ37lw8PDxo06YNAMWLFwcgLi4OJyenbNd076Z48eIMGjTIuOXH\n29ubGzdusGTJEvbs2cOUKVMeqB0REXm83TdoX3/9dXbt2kVISAhvvPHGHeu0b9+elJQUVq1axVdf\nfUWJEiWoVasWq1evpk+fPuzdu5dOnTpRqFAhlixZwtSpU5k7dy7nz5+nSpUqjB8//q5t29raEhoa\nymuvvUZISAhLliyhTZs2ODg4MHv2bIYNGwbcGpmGh4dbLVAaNmwYZcuWZcWKFaxatYpnn32WadOm\n0bNnz/uemJIlS7Jq1SqmTZvGvHnzSEtLo0KFCvTu3Zv+/fsb11Cfe+45Xn31Vb788kt27NjBN998\nc9+2swQGBuLk5MSCBQvYsGED9vb2WCwWZs+eTbNmzR64HREReXzZZOpp+PK/slYdX3tlJBQple3z\npIJvPnhjY87lYs9ERB5fWb87Y2NjcXZ2zva5HsEoIiJiIgWtiIiIiRS0IiIiJlLQioiImEhBKyIi\nYiIFrYiIiIkUtCIiIiZS0IqIiJgoV14qIE+XXcOb3/Gma9BDKEREHpZGtCIiIiZS0IqIiJhIQSsi\nImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQioiImEhBKyIiYiIFrYiI\niIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIi\nJlLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiY\nKF9ed0AeP43Dt0ORUnndjRxJKvhmXnfhyTbmXF73QOSpoxGtiIiIiRS0IiIiJlLQioiImEhBKyIi\nYiIFrYiIiIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgXtI4iL\ni6Nr1668+OKL1K1bl4CAABISEozPLRYL8+fP58MPP6RRo0bUrl2bXr16cezYMb755htee+013N3d\nad26NVu2bLFq+8SJEwwdOpT69evj4eHBwIEDiYmJwWKxsGfPnn/6UEVEJIf09p4cWrRoEWFhYfj4\n+BAWFsbVq1dZsGAB3bt3JzIyEi8vLwBmzZqFp6cnYWFhJCcnM3HiRAIDA0lPTyc4OJjixYszffp0\n3nnnHbZs2UK5cuW4ePEib775JtevX2fo0KGUKVOGtWvX8v777+e4vx06dLhvnfT09By3LyIid6ag\nzYELFy4wdepUGjVqRGRkpFHeokUL2rRpw/jx442gLV68ODNmzCBfvlun+ptvvmH79u3MmzePJk2a\nAJAvXz6CgoI4cOAA5cqVY8mSJZw4cYJ169ZRq1YtAHx8fOjatSvff//9P3y0IiLyKBS0OZCYmMiV\nK1d47bXXrMoLFChA+/btiYyM5OjRowB4eHgYIQtQtmxZAF544QWjrGTJkgCcP38egF27duHi4mKE\nbBZfX98cB210dPR96yQnJ+Pj45Oj9kVE5M50jTYH0tLSgP+G5u0cHR2BW6NegCJFityxjcKFCxvf\n29jYWH12+vRpSpcunW2bO+1PREQebwraHChRogQAp06dyvZZSkoK8N9Rak44ODiQmpqarfz06dM5\nblNERPKGgjYHPDw8KFSoEJ999plVeXp6Ops2baJy5cqUL18+x+03btyYpKQkfvrpJ6vymJiYHLcp\nIiJ5Q0GbA8WKFWPQoEHs3r2b4OBgtm7dSkxMDF26dOHEiROEhIQ8Uvs9evTAycmJfv36sXr1arZv\n387QoUOJj48HwNZW/2wiIk8K/cbOoYCAAKZMmcKpU6cYOnQoo0ePxsHBgSVLljzygiIHBweWLl2K\nh4cH4eHhDB06lIyMDIKDgwHr67siIvJ406rjR9C2bVvatm17188PHTqUrWzcuHGMGzfOqqxOnTpW\ndRMSEjh16hQRERFW9UJDQ7Gzs6NKlSqP2HMREfmnKGgfQykpKbz99tvEx8fz8ssvY2try969e1m+\nfDn+/v4ULVo0r7soIiIPSEH7GGrTpg1Xrlxh1apVxMTEkJGRQZUqVRg+fDhdu3bN6+6JiMhDUNA+\npjp27EjHjh3zuhsiIvKItBhKRETERApaEREREyloRURETKSgFRERMZGCVkRExERadSzZ7BreHGdn\n57zuRg6dy+sOiIhY0YhWRETERApaEREREyloRURETKSgFRERMZGCVkRExEQKWhERERMpaEVEREyk\noBURETGRglZERMRECloRERETKWhFRERMpKAVERExkYJWRETERApaEREREyloRURETKSgFRERMZGC\nVkRExEQKWhERERMpaEVEREykoBURETGRglZERMRECloRERETKWhFRERMpKAVERExkYJWRETERApa\nEREREyloRURETKSgFRERMZGCVkRExEQKWhERERPly+sOyOOncfh2KFIqr7thuqSCb+Z1F54sY87l\ndQ9Enkga0YqIiJhIQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQ\nioiImEhB+7+8vb0JDAzM9v2D2rNnDxaLhZiYGDO6B+SsXyIikrf0CMY7mD59OgULFszrboiIyFNA\nQXsHbm5ued0FERF5Smjq+A7+PkVrsViYN28eU6ZMoWnTpri6uuLr68u2bdvu2c6MGTOwWCzMnj3b\nKDt69ChDhw6lfv36uLm50aFDB7Zs2ZJt25iYGHx9fXF3d6dly5Zs3bo19w5QRET+MRrRPqCoqCie\nf/55Ro4ciY2NDZ988gmDBg1i69atODk5Zas/Y8YMIiIieO+99+jZsycAycnJvPHGGxQpUoR3332X\nkiVLsnHjRt566y0mTpxIhw4dAIiOjua9997jlVdeYdCgQZw4cYJRo0Zx/vx5qlatmqP+Z7V9L+np\n6TlqW0RE7k5B+4CKFy/OvHnzyJfv1ikrWrQoPXv2ZNu2bbz5pvXr1mbMmMGMGTMYO3Ysfn5+VuVX\nrlxh3bp1VKxYEQAfHx+Cg4MJDw+nbdu22NvbM336dNzc3IiIiDC2rVatGt26dfsHjlRERHKTgvYB\neXh4GCEL4OzsDMClS5es6i1atIgff/yRxo0bW4UswK5du3Bzc6NcuXJkZGQY5a1atSI2Npb9+/dT\npkwZTp48SdeuXa22ffHFFylfvnyO+x8dHX3fOsnJyfj4+OR4HyIikp2C9gEVKlTI6mcbGxsAMjMz\nrcp/+uknmjVrxo4dO9i2bRve3t7GZ2fPniU1NZVatWrdcR8nT57Ezs4OgDJlymT7vGzZso90DCIi\n8s9T0Oayd999ly5dutCxY0c++OADPDw8KFmyJHBr+rlGjRoMHTr0jts6Oztz/vx5AFJTU7N9fubM\nGYoVK2Ze50VEJNdp1XEuc3R0JH/+/ISFhXH27FnGjh1rfNawYUMOHz5M1apVcXNzM77279/PtGnT\nuHr1KlWqVKFy5cps2rSJmzdvGtv+/PPPJCcn58UhiYjII1DQmsTV1ZWAgAC++OILNm/eDMDAgQO5\ncuUKXbt2Zf369ezevZupU6cSGhpKvnz5jGuwISEh/P777/Tp04etW7eydu1a+vfvr4doiIg8gRS0\nJnrrrbdwcXFh7NixpKam8swzz7BmzRqeeeYZJk6cSP/+/fnyyy/p378/n3zyibGdj48Ps2fP5vz5\n87zzzjvMnDmTfv36UbNmzTw8GhERyQmbzL+v5pH/s7JWHV97ZSQUKZXX3TFdUsE3719J/mvMubzu\ngchjKet3Z2xsrHFHyu00ohURETGRglZERMRECloRERETKWhFRERMpKAVERExkYJWRETERApaERER\nEyloRURETKSXCkg2u4Y3v+NN108fPYBBRMynEa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiI\niRS0IiIiJlLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS0IqIiJhIQSsiImIiBa2IiIiJFLQiIiIm\nUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJlLQioiImEhBKyIiYiIFrYiIiIkUtCIiIiZS0IqIiJhI\nQSsiImIiBa2IiIiJFLQiIiImUtCKiIiYSEErIiJiIgWtiIiIiRS0IiIiJsqX1x2Qx0/j8O1QpFRe\nd0Mec0kF38zrLojknjHnTGtaI1oRERETKWhFRERMpKAVERExkYJWRETERApaEREREyloRURETKSg\nFRERMZGCVkRExERPddBmZmbmdRdEROT/uKcyaNPT05k8eTKLFi36R/drsVj44IMPTGs/JCQENzc3\n09oXEZHc91QG7alTp5g7dy7p6el53RUREfk/7qkMWhERkcfFYxW03t7eBAYGWpWlpqZisViIiIgg\nOTkZi8XC+vXrGT16NA0aNMDd3R1/f3/+85//ALBnzx58fHwA+Pjjj7FYLABERETg6enJmjVraNKk\nCZ6enuzcuROAzZs307VrVzw9PXF1daVZs2aMGzeOixcvWvVl+/btdOnSBQ8PD1566SUGDBjAH3/8\ncdfj+eabb3Bzc6N///7G6Pry5cuEh4fTvHlzXF1dadGiBXPmzOHGjRtW2x48eJCAgAA8PDxo2LAh\nM2fO1DVnEZEn0BP59p6wsDAaNmxIWFgYFy9eZPLkyQQFBbFjxw6ef/55pk+fzuDBg/Hz86Ndu3bG\ndpcvXyYyMpIPPviAtLQ0PD09WbduHe+//z5+fn707duXzMxMtm/fzrJlyyhQoADDhw8HYP369Qwf\nPpzGjRsTHh5ORkYGERERdO/enY0bN1KqlPXbbr755huCg4Px9vZm8uTJ5M+fn+vXrxMQEMChQ4fo\n378/NWrUICEhgenTp5OUlMTEiRMBOHz4MF26dKFy5cqEhoYCEBUVxe+//46Njc0/dJZFRCQ3PJFB\nW61aNaZNm2b8fO3aNUaMGMHevXvx8vLC1dUVgIoVK+Lp6WnUu3HjBoMHD6ZFixZG2W+//Yavry9j\nx441yry8vNi3bx/ffvst8P/bu++gqM7uD+BfWDCUFUbUKBEHRLOKSxMpCqhsYNYy4mAcC2qCRAOW\nmLHmdZRExALqWIg9sYDdjF3BiooakMDoWMaIZURRFFGQDrLs+f3BjzteFwLm5b4Uz2dm/9jnnr33\nPM+j9+xtS9Xdy6tXr0bPnj2xdetWodj16NEDwcHBSEtLg1qtFj5fXWSHDBmCZcuWQSaTAQBOnDiB\nGzduYO3atRg8eLCwLQsLC0RGRiIwMBCOjo7YuHEj9PX1ERMTg7Zt2wIAvL294efnh9LS0n81Zl9/\n/XWdMXxNmzHGGl6zLLRubm6i9506dQIAFBcX1/nZ6lPJ1ebNmwegqlg/efIEmZmZSE9PR25uLszN\nzQEAGRkZyM7Oxvjx40VHlLa2tsLp52o3b97EkSNHYGxsjPDwcKHIAsCff/4JAwMDqFQqaDQaoX3Q\noEGIjIzE5cuX4ejoiJSUFLi6ugpFFgDMzMzg4+ODU6dO1dlHxhhjTUezLLQmJiai9/r6VZea63MN\ns127dqL32dnZWLp0KSGQn+gAABFRSURBVBISEqDVamFlZQV7e3sYGRkJ68vLywMAtG/fvs7137t3\nD76+vrhw4QKio6OFQg4Aubm50Gg0cHJyqvGz2dnZAIC3b9/q5AkAn3/+eZ3br83hw4frjHn27Jlw\nfZsxxljDaHKFVqvVit5/eENSQyIiTJw4ESUlJdixYwccHR1hZGQEABgxYoRwhGxmZgYAeP36tc46\nrl69CisrK9jY2ACouqFr48aNCA8PR2xsLPz8/ITT12ZmZjA3N8e2bdtqzKdNmzYAAAsLC7x69Upn\neW5u7n/XYcYYY/9zTequY7lcjqysLFFbSkrKR6/n/dO1/yQvLw8PHjzAkCFD4O7uLhTZZ8+eIT09\nXTiitbW1Rfv27XHmzBnR558/f47vv/8ep0+fFtqqj3rnzJmDjh07Yt68eSgpKQEAeHp6Ij8/H3p6\nenBwcBBeGo0GK1aswKNHjwBUXY9NTU3FixcvhPWWlZXhypUrHz0WjDHGGleTKrR+fn7IyMjAkiVL\nkJSUhK1bt2LDhg1o1arVR62ndevW0NPTQ0pKCpKTk2s9pWxhYQFra2scPnwYhw4dQnJyMmJjYxEY\nGAiNRiMUSH19fcyZMwe3b9/GlClTcP78ecTFxSE0NBQdO3bEqFGjdNYtl8sRERGBzMxMLF++HAAQ\nEBAApVKJ0NBQxMTEICkpCXv37sXUqVPx7NkzODo6AgB++OEHGBkZISgoCMeOHcPZs2cRHByMwsLC\njxoHxhhjja9JnToODQ1FQUEB4uPj8ccff8DR0RGbNm3Seba2LnK5HJMmTRKKWFxcXK2xmzZtQmRk\nJKKiolBZWYlOnTph7NixkMlkWLVqFZ48eQJra2sEBARALpdjy5YtmDVrFuRyOfr27YuZM2fqPNpT\nrV+/fhg+fDj2798PtVoNLy8v7Ny5E+vWrUNsbCxycnJgYWEBHx8f/Pjjj8Kp4y+++AL79+/HihUr\nEBERAZlMhqFDh8LZ2Rm7d+/+qLFgjDHWuPSIfwWB/b/qm6HK1WGAac1fHhirlmE0trFTYKzhhOf/\n649W7zsTEhJgZWWls7xJnTpmjDHGWhoutIwxxpiEuNAyxhhjEuJCyxhjjEmICy1jjDEmIS60jDHG\nmIS40DLGGGMS4kLLGGOMSahJ/TIUaxqu/kdV40PXjIn9+wf8GfuU8BEtY4wxJiEutIwxxpiEuNAy\nxhhjEuJCyxhjjEmICy1jjDEmIS60jDHGmIS40DLGGGMS4kLLGGOMSYgLLWOMMSYhLrSMMcaYhLjQ\nMsYYYxLiQssYY4xJiAstY4wxJiEutIwxxpiE+M/kMUFlZSUA4OXLl42cCWOMNR/V+8zqfeiHuNAy\nQU5ODgBg3LhxjZwJY4w1Pzk5ObC2ttZp1yMiaoR8WBNUVlaGO3fuoH379pDJZI2dDmOMNQuVlZXI\nycmBvb09jIyMdJZzoWWMMcYkxDdDMcYYYxLiQssYY4xJiAstY4wxJiEutIwxxpiEuNAyxhhjEuJC\nyxhjjEmICy1jjDEmIS60jDHGmIS40DIAwI0bN/DNN9/AxcUFHh4emDt3rvCTjI3t7NmzGDNmDFxd\nXeHp6YlJkybh9u3bopijR4/C398fjo6O8PHxwdq1a/Hu3TtRzLt377BmzRqoVCo4ODjA398fx44d\n09lednY2Zs+ejb59+8LZ2Rnffvstbt26pRMn5Zjt27cP3bt3R0pKiqg9MTERI0eOhLOzM7y8vLB4\n8WIUFRWJYogI27dvh1qthoODA9RqNWJiYvDhb9MUFhYiPDwc/fr1g5OTE0aOHInLly/r5PLo0SNM\nnjwZ7u7ucHFxwZQpU5CRkfGv+/bw4UNMmzYNbm5ucHFxQWBgIJKTk0UxLWU+k5KSEBgYiN69e8Pb\n2xuzZ8/GixcvRDHNeU41Gg3GjBmDX375RWdZU+5XfXJrUMQ+eXfv3iUnJycaP348nT17lvbt20ce\nHh40ZMgQKi8vb9TcDh06RAqFgmbNmkUJCQkUFxdHI0aMIKVSSampqUREdODAAVIoFBQWFkYXL16k\ntWvXkp2dHc2fP1+0rlmzZpG9vT1t2bKFLly4QDNnziSFQkFHjx4VYoqLi0mtVtOAAQPo8OHDFB8f\nT8OHDydnZ2d6+PChECflmD18+JCcnJxIoVDQtWvXhPbExESys7OjadOmUUJCAm3bto2cnJwoKChI\n9PnVq1dTjx49aOXKlXTx4kWKiIgghUJBGzZsEGK0Wi0FBgaSq6sr7d69m86dO0fBwcFkZ2cn2ubL\nly+pT58+NHToUIqLi6OjR4+Sr68veXl50Zs3bz66b48ePaLevXtTQEAAxcfH07lz52jcuHGkVCrp\nxo0bRNRy5vOvv/4iOzs7Gj9+PCUkJNCRI0dIpVKRr68vFRQUEFHzntPi4mKaOnUqKRQK+vnnn0XL\nmnK/6ptbQ+JCyyg0NJT69+9PZWVlQtvNmzdJoVDQ3r17GzEzov79++v8BygsLCR3d3cKCQmhsrIy\n8vDwoOnTp4tifv/9d+revbuwM719+zYpFAratWuXKC4kJIS8vb2psrKSiIi2bdtGCoWC0tPTRdvz\n8vKiGTNmCG1SjVl5eTkFBASQj4+PTqEdOnQojRgxgrRardAWHx9PCoWCEhMTiahqZ6NUKikqKkq0\n3vDwcHJwcKD8/HwiIjp9+jQpFAq6cOGCEFNRUUHDhg2jkSNHCm0RERHk5OREr1+/FtqysrLIwcGB\nVq5c+dH9CwkJIZVKRUVFRUJbcXEx+fn50erVq1vUfM6ePZt69epFxcXFQltSUhIpFAo6cuQIETXf\nOb106RINHDiQ3N3dayy0Tblf9cmtofGp40/cu3fvkJSUBF9fX3z22WdCu6OjI2xsbJCQkNBouRUX\nF+Orr77C2LFjRe1yuRyWlpZ4+fIlbt26hby8PAwZMkQU4+/vDyIS8r906RIA6MQNHToUr169Ek5F\nX7p0CV9++SUUCoVoeyqVCpcuXYJWq5V0zNasWYOSkhJMnTpV1J6VlYX79+9j8ODB0NPTE9r9/Pxg\nYmIibDM5ORkVFRU1jkd5eTmuXr0q9FMul6N///5CjIGBAQYPHoybN2/i9evXQpyHhwfatm0rxFla\nWsLV1fWj+1lQUIArV65gzJgxMDU1FdpNTExw7tw5zJw5s0XNZ3l5OQwNDWFsbCy0VY/j27dvm+2c\nFhQUIDQ0FN27d8fx48d1ljflftU3t4bGhfYTl5mZifLycnTt2lVnWZcuXfDw4cNGyKqKqakpFi5c\nCLVaLWp//PgxHjx4ADs7OyG/D/Pv0KEDTExMhOWPHj1CmzZtYGFhIYrr0qULAODBgwdCnK2trU4u\nXbp0QUlJCZ4/fy7ZmCUlJWHXrl1YuXIlTExMRMtq66ehoSGsrKyE5bXF1dRPGxsbnb/S9H5ceXk5\nnj17Vut4ZGRkQKPR1Lt/9+7dQ2VlJaysrLBkyRJ4enpCqVRi1KhRSEtL+8f8m+N8jh8/HiUlJVi+\nfDlyc3ORmZmJZcuWwdzcHAMHDmy2c2pkZIS4uDhER0ejQ4cOOsubcr/qm1tD40L7iSssLARQ9S3/\nQ6ampsLypqKoqAizZ8+GoaEhQkJChPxat26tEyuXy4XlBQUFtcZUr7c+cYWFhZKMWV5eHn766SdM\nnjwZjo6OOsvru82CggLIZDKdQv1+/tVxdY1HQUEBgNrHVqvVori4uN59rD7yiIyMxJMnTxAZGYno\n6GhotVpMmDABN27caDHzCQAeHh6YMWMGYmNj0bdvX/j5+eHOnTvYvHkzLC0tm+2ctmrVqsYvJdWa\ncr8aa3/Hf/j9E6fVagFAdBrlffr6Tee7WHZ2NiZPnowHDx4gOjoatra29c6fiGqNef/zdcXp6+tL\nMmYLFiyAlZUVpkyZUuPy/7af1W0fMx51bfP99dVHRUUFAKBdu3bYvHmzcITi7u4OtVqNX3/9FR4e\nHv+4zeYynwCwePFi7N69G6NHj8agQYNQWlqKvXv3YuLEidiwYUOLmNOaNOV+Ndb+runsRVmjMDc3\nB4Aav8kVFxfX+A2xMVy/fh0jRozAkydPsHHjRvj6+gIAzMzMANScf1FRkZC/mZlZrTEA6h0nl8sb\nfMz279+P5ORkLFu2DEQEjUYj7BC0Wi00Go3Qz5oeQXh/m2ZmZtBoNCgtLRXFfHikWJ/xqGts9fT0\nRNda61J9FOHj4yM6DWhmZgYXFxfcuXOnRcwnUPWlcM+ePQgICEBERAQ8PT3h6+uL3377DV27dkVY\nWFiLmNOaNOV+1Te3hsaF9hPXuXNnGBoa1vis2ePHj9GtW7f/fVIfOHnyJIKCgmBoaIi9e/eKboqo\nPoX1Yf7Z2dkoKSkR8u/atStyc3ORn58vinv8+DEAiOJqGwtTU1NYWlo2+JjFxcWhpKQEgwcPhlKp\nhFKpxJw5cwAAEyZMgFKpFPpZnW+1iooKZGZmivIHdMejpn4+ffpU53nF9+OMjY3RqVOnWvvZtWvX\njzoCqL6mVl5errPs3bt3MDIyahHzCVTddENEcHNzE7XLZDK4urri+fPnwvXN5jynNWnK/1brm1tD\n40L7iWvVqhU8PT2RkJCAsrIyof3WrVvIyMjAgAEDGjE74MyZM5g7dy569OiBgwcPokePHqLlvXr1\ngrm5OU6ePClqP3HiBPT09ISiXN2PuLg4nbj27dujZ8+eQlx6erpwIwZQ9e33woUL8Pb2hkwma/Ax\nW7RoEQ4ePCh6zZw5U7TMysoK3bp1w6lTp4SjXQA4f/48SktLhW3269cPMpmsxvEwMjISTs0OGDAA\n+fn5oof+NRoN4uPjYW9vL9y5OWDAACQnJ+PNmzdC3IsXL5Camir6wlMftra2sLGxwenTp0Xj9vr1\na1y/fh0eHh4tYj4BwMbGBgYGBjo/OFJZWYm0tDS0bdsW3bt3b/ZzWpOm/G+1vrk1NFl4eHi4JGtm\nzYaNjQ127tyJpKQktG7dGtevX8f8+fNhZWWFRYsWwcCgcS7lv337FhMmTICBgQHCwsJQUlKCrKws\n4ZWXlwdLS0u0atUKO3bswPPnz2FgYIDjx49j3bp1GD58OEaNGgWg6q7V9PR07Nq1C0DVznbdunU4\nd+4cFi5cKOyYFQoFTpw4gcOHD8PMzAxPnz5FWFgYcnJysGrVKuEu14YcszZt2qBDhw6iV3Z2Ns6c\nOYPvvvsOLi4uQh+2b9+Ou3fvwtjYGImJiVi6dCnc3NwwY8YMAFU3dOTn5yMmJgZFRUXQaDTYtWsX\n9uzZg6lTpwo7EltbWyQmJmL//v0wNjZGTk4Oli1bhtu3byMqKgqdO3cWxuPAgQM4c+YMzM3NkZ6e\njnnz5sHAwAArVqzQuZGlLtbW1jhw4ACSk5OF9S1YsACFhYVYvXo12rZt2+znEwCMjY2h0Wiwe/du\nZGVlwdDQEPfv30dUVBTS0tKwcOFC2NnZtYg5Xb9+Pezt7aFSqYS2ptyv+uTW4CR5Opc1O9euXaPR\no0eTg4MD9enTh+bOnUuvXr1q1JyOHz9OCoWi1tfAgQOF2D179tDAgQNJqVSSSqWi6OhoqqioEK2v\ntLSUli5dSl5eXuTg4ED+/v504sQJne1mZmbS9OnTycXFhXr16kVBQUF08+ZNnTgpx+zkyZM6P1hB\nVPXw/rBhw8je3p68vb0pIiKCCgsLRTEajYbWr19PKpWK7O3tSa1WU2xsrM42cnNzad68eeTu7k5O\nTk40atQounLlik7c33//TcHBweTk5ERubm40ZcoUevz48b/uW2pqKgUFBZGzszO5uLhQSEgI3b9/\nXxTTUubz4MGDFBAQQPb29uTh4UFBQUGUkpIiimnuc1rTD1Y09X7VJ7eGpEf0wYlvxhhjjDUYvkbL\nGGOMSYgLLWOMMSYhLrSMMcaYhLjQMsYYYxLiQssYY4xJiAstY4wxJiEutIwxxpiEuNAyxhhjEuJC\nyxhjjEno/wAVZfVKm40yaQAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a13133790>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "uniqueCount_percentage('first_affiliate_tracked', -1)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "collapsed": true
   },
   "source": [
    "The first_affiliate_tracked means that before signing up, what is the first marketing the user is interacted with. It shows that the aside from the untracked information, the OMG group (a marketing company has the most affect with highest user interacting number). Another popular company the user is interacting is NAN group.\n",
    "\n",
    "This shows that when the user are interacting with omg and nan, they have more potential to log into the airbnb."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Data Analysis of Affiliate Related Stuffs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Based on the data visulization we obtained in the former chapter, we can put the two assumptions here:\n",
    "   1. A large scale of customer sign up and make the booking directly without any optimization methods, this means that they have strong motivation and goal to start their trip using airbnb.\n",
    "   2. Using Google with omg/nan might be the most common way user make booking aside from the direct way.\n",
    "   \n",
    "Next thing we want to do check the age composition in the differemt choices"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Age composition between Google and Direct"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "def box_hist(target, keyword):\n",
    "    '''\n",
    "    This function draws the box plot of data with given keyword equals the target one\n",
    "    *param target: the target value which the data[keyword] should be equal to\n",
    "    *param keyword: the keyword of data that you are interested in\n",
    "    '''\n",
    "    age_group  = train_users[['country_destination','affiliate_channel','affiliate_provider','first_affiliate_tracked','age']]\n",
    "    age_group = age_group[age_group['age']>=60]\n",
    "    age_group = age_group[age_group['age']<=80]\n",
    "    df_3 = age_group[age_group[keyword]==target]\n",
    "    df_3 = df_3['age']\n",
    "    plt.figure()\n",
    "    df_3.plot.hist(stacked=True, bins=20)\n",
    "    plt.figure()\n",
    "    df_3.plot.box()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaUAAAEMCAYAAABkwamIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a131d9610>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEMCAYAAADAqxFbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAFlFJREFUeJzt3X1UlHXex/HP8PwkpKboWUpAUlux\n0AZZyUXMB0xjVTyZiRontaOipquVm26d1ERW0122cHIzfKK2TnralVq7jWrtrHtczS0dH/CwRsK6\nmqyiuIggzv2Hd3PvL6yZyRmRer/O4Q+u63fNfK9/eHPNNQwWh8PhEAAA/8evpQcAANxcCAMAwEAY\nAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAMhAEAYAho\n6QGuR319vex2uzp06CB/f/+WHgcAWoWmpiadPn1aiYmJCgkJaba/VYfBbrcrOzu7pccAgFapuLhY\nVqu12fZWHYYOHTpIunpynTp1auFpAKB1OHnypLKzs50/Q7+uVYfhq5ePOnXqpJiYmBaeBgBal296\nCZ6bzwAAg9th2LVrlx5++GHdc8896t+/v+bNm6d//etfxpo///nPevDBB5WUlKR7771XS5Ys0YUL\nF7w+NADAd9wKw549ezRlyhQFBARoxYoVmj9/vv7+979r4sSJqq2tlSTt3LlT06dPV3R0tFatWqXJ\nkydry5Ytmjlzpk9PAADgXW7dY3jjjTcUEhKil19+WWFhYZKk6Oho5eTkqLS0VKNGjdKKFSv04x//\nWL/97W9lsVgkSZ07d9acOXO0c+dOpaWl+e4sAABe49YVw6VLlxQYGKjQ0FDntvbt20uSampqdOLE\nCR09elT333+/MwqSNHjwYIWFham0tNTLYwMAfMWtMEyYMEF1dXXKz8/XmTNnVFlZqWXLlikqKkoZ\nGRkqLy+XJHXt2tU4LjAwUDExMc79AICbn1svJaWkpGjOnDlauXKlioqKJElt2rTR2rVr1blzZ+3b\nt0+SFBER0ezY8PBw530IT2RlZblc09DQ4PHjAgC+nVthWLJkiTZv3qyHHnpIw4YN08WLF/Xaa69p\n8uTJeumll3TlyhVJMl5G+m9+frwrFgBaC5dhOHXqlIqLizVq1CgtXrzYuT09PV0PPfSQFi1apGef\nfVaSrvnW1P/85z+65ZZbPB5s69atLtdUVVVp0KBBHj82AOCbufxV/sSJE3I4HEpOTja2+/v7y2q1\n6p///Keio6MlSZ9//rmxprGxUZWVlUpISPDiyAAAX3IZhtjYWAUEBGj37t3G9qamJu3du1ft27dX\n9+7dlZCQoD/96U/Ol5Uk6f3339fFixc1YMAA708OAPAJly8ltW3bVo899pgKCwsVEBCgYcOG6dKl\nS3rjjTd04MABLV++XBaLRXPnzlVubq5mzJihBx98UMePH9dvfvMb9evXT+np6TfgVAAA3uDWzefH\nH39cMTEx2rx5s0pKShQeHq4ePXpo06ZN6tu3r6Srf7NQUFCgwsJCzZkzR7fccovGjBmjuXPn+vQE\nAADe5fanq44ZM0Zjxoz51jUZGRnKyMi47qEAAC2H95ECAAyEAQBgIAwAAANhANyUmJgoi8Xi06/E\nxMSWPk2AMADustvtcjgcbn91earEo/UOh0N2u72lTxMgDAAAE2EAABgIAwDAQBgAAAbCAAAwEAYA\ngIEwAAAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBgIAwAAANhAAAYCAMA\nwEAYAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAMhAEA\nYAhwtaCqqkqDBg361jVlZWWqqalRSkrKNfcXFRUpNTX1u00IALihXIahY8eOKi4ubrb9wIEDWr58\nubKzsyVJhw4dkiQtXbpUcXFxxtru3bt7Y1YAwA3gMgxBQUGyWq3GtpqaGs2fP19Wq1VPP/20JMlu\nt8vf318PPPCAQkNDfTMtAMDnXIbhWlauXKkzZ85o/fr1Cgi4+hCHDh1SfHw8UQCAVs7jMNjtdr31\n1luaOXOmYmNjndsPHjyoiIgITZkyRfv27dOVK1f0k5/8RE888YS6du3q8WBZWVku1zQ0NHj8uACA\nb+fxu5JsNpuioqKUk5Pj3FZbW6vKykpVVFQoLS1Na9eu1cKFC1VWVqZx48bp+PHj3pwZAOBDHl0x\nVFZWqrS0VDNmzFBERIRze1BQkIqKitShQwclJCRIkqxWq6xWqzIzM7VmzRrl5eV5NNjWrVtdrnHn\nHVMAAM94FIZ3331XDodDo0ePNrYHBwerX79+zdbHxcUpPj5ehw8fvr4pAQA3jEcvJe3YsUO9e/dW\nTEyMsb2iokLFxcU6efJks2Pq6+vVrl2765sSAHDDuB2G+vp6HT58WMnJyc32nTt3TosXL9brr79u\nbP/kk090/Pjxa15NAABuTm6/lHT06FFdvnxZ3bp1a7bvrrvu0uDBg/XKK69IkpKTk1VeXq7CwkIl\nJCRo0qRJ3psYAOBTboehurpakhQZGdlsn8Vi0QsvvKB169Zp27ZtevXVVxUVFaURI0Zo9uzZCg4O\n9t7EAACfcjsM9913n8rKyr5xf0hIiHJzc5Wbm+uVwQAALYNPVwUAGAgDAMBAGAAABsIAADAQBgCA\ngTAAAAyEAQBgIAwAAANhAAAYCAMAwEAYAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDA\nQBgAAAbCAAAwEAYAgIEwAAAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBg\nIAwAAANhAAAYCAMAwEAYAAAGwgAAMAS4WlBVVaVBgwZ965qysjJJ0ttvv61169bpiy++ULt27TRq\n1CjNmDFDQUFB3pkWAOBzLsPQsWNHFRcXN9t+4MABLV++XNnZ2ZKkN998U7/85S81duxYzZs3T599\n9plefvllnT59Ws8//7z3JwcA+ITLMAQFBclqtRrbampqNH/+fFmtVj399NO6dOmSVq1apYyMDC1Z\nskSSlJ6ervDwcK1cuVKPPvqounbt6pszAAB41Xe6x7By5UqdOXNGzz//vAICArR//36dPXtWw4cP\nN9ZlZmbK4XCotLTUK8MCAHzP4zDY7Xa99dZbeuyxxxQbGytJKi8vl6RmVwXR0dEKCwtz7gcA3Pxc\nvpT0dTabTVFRUcrJyXFuq62tlSS1adOm2fqIiAjnfk9kZWW5XNPQ0ODx4wIAvp1HVwyVlZUqLS3V\nhAkTFBER4dx+5coVSZLFYrn2k/jxrlgAaC08umJ499135XA4NHr0aGN7ZGSkpKtXDtHR0ca+Cxcu\nXPNKwpWtW7e6XOPOW2kBAJ7x6Ff5HTt2qHfv3oqJiTG2f3VvoaKiwth+6tQp1dXVKSEh4fqmBADc\nMG6Hob6+XocPH1ZycnKzfb1791ZUVJRKSkqM7du2bZPFYlFaWtr1TwoAuCHcfinp6NGjunz5srp1\n69ZsX1BQkHJzc7Vs2TKFhIRo2LBh2r9/v2w2m7Kysq55DADg5uR2GKqrqyX9//2Er3vkkUcUGBio\njRs3qqSkRB07dtS0adM0Y8YM70wKALgh3A7Dfffd5/xMpG8yfvx4jR8//rqHAgC0HN5HCgAwEAYA\ngIEwAAAMHn8kBvB9cPdz/6NzFxt9/jyxC97x+XNEhQbqs2eH+vx58MNBGPCDdO5ioyqWj2jpMbzi\nRsQHPyy8lAQAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAMhAEA\nYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBgIAwAAANhAAAYCAMAwEAYAAAGwgAA\nMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAEODuwvLycq1evVp/+9vf1NTUpO7du2v27Nnq16+f\nJKmmpkYpKSnXPLaoqEipqanemRgA4FNuheHYsWMaN26cbrvtNi1evFiBgYFav369pk6dqs2bNysp\nKUmHDh2SJC1dulRxcXHG8d27d/f+5AAAn3ArDPn5+YqMjNTmzZsVHh4uSUpNTdXIkSP14YcfKikp\nSXa7Xf7+/nrggQcUGhrq06EBAL7jMgznz5/Xxx9/rDlz5jijIElhYWHasWOH8/tDhw4pPj6eKABA\nK+fy5vORI0fU1NSkmJgYLV26VKmpqerZs6fGjh2rvXv3OtcdPHhQgYGBmjJlivr06aOkpCRNmzZN\n//jHP3x6AgAA73J5xVBdXS1JysvLU48ePZSXl6fGxkbZbDbl5ORo06ZNSkhIUGVlpUJDQzV69GhN\nmzZNn3/+uQoLCzVu3Dht2bJFt99+u0eDZWVluVzT0NDg0WMCAFxzGYbGxkZJ0q233iqbzSZ/f39J\nUt++fTV06FAVFBTIZrOpqKhIHTp0UEJCgiTJarXKarUqMzNTa9asUV5eng9PA/BMWNxq9dqwoKXH\n8IqwuGhJI1p6DHyPuAxDRESEJCk9Pd0ZBUmKjIxUnz59tGfPHgUHBzvftvrf4uLiFB8fr8OHD3s8\n2NatW12uqaqq0qBBgzx+bKDu87mqWP79+GEau+Cdlh4B3zMu7zF89dbTS5cuNdvX0NCgkJAQVVRU\nqLi4WCdPnmy2pr6+Xu3atfPCqACAG8FlGOLj4xUbG6vt27ervr7eub26ulr79u1TSkqKzp07p8WL\nF+v11183jv3kk090/Pjxa15NAABuTm79HcPChQs1ffp05eTkaPLkyWpsbFRhYaEkadasWbr99ts1\nePBgvfLKK5Kk5ORklZeXq7CwUAkJCZo0aZLvzgAA4FVuhSEtLU0bNmzQiy++qCeffFJ+fn6yWq1a\nvXq1unTpIkl64YUXtG7dOm3btk2vvvqqoqKiNGLECM2ePVvBwcE+PQkAgPe4/VlJVqtV69ev/8b9\nISEhys3NVW5urjfmAgC0ED5dFQBgIAwAAANhAAAYCAMAwEAYAAAGwgAAMBAGAICBMAAADIQBAGAg\nDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQ\nBgCAIaClBwBaSuyCd1p6BK+ICg1s6RHwPUMY8INUsXyEz58jdsE7N+R5AG/jpSQAgIEwAAAMhAEA\nYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBgIAwAAIPbYSgvL1dubq6Sk5PVp08f\nPfzww/rrX/9qrHn77beVmZmpu+66S+np6fr1r3+thoYGrw8NAPAdt8Jw7NgxjRs3TidOnNDixYv1\nq1/9Sv7+/po6dao+/fRTSdKbb76pp556SklJSSooKNDo0aO1du1aPffccz49AQCAd7n1sdv5+fmK\njIzU5s2bFR4eLklKTU3VyJEj9eGHH+rOO+/UqlWrlJGRoSVLlkiS0tPTFR4erpUrV+rRRx9V165d\nfXcWAACvcXnFcP78eX388ccaN26cMwqSFBYWph07dmju3Lnav3+/zp49q+HDhxvHZmZmyuFwqLS0\n1PuTAwB8wmUYjhw5oqamJsXExGjp0qVKTU1Vz549NXbsWO3du1fS1fsPkppdFURHRyssLMy5HwBw\n83P5UlJ1dbUkKS8vTz169FBeXp4aGxtls9mUk5OjTZs2qba2VpLUpk2bZsdHREQ493siKyvL5Rpu\nbAOA97kMQ2NjoyTp1ltvlc1mk7+/vySpb9++Gjp0qAoKCpSSkiJJslgs13wMPz/eFYvWLzExUQcP\nHvToGEu+Z8/Rs2dP2e12zw4CvMxlGCIiIiRdvZn8VRQkKTIyUn369NGePXs0ZMgQSVJtba2io6ON\n4y9cuHDNKwlXtm7d6nJNVVWVBg0a5PFjA98FP7DxQ+HyV/m4uDhJ0qVLl5rta2hoUEhIiPPeQkVF\nhbH/1KlTqqurU0JCghdGBQDcCC7DEB8fr9jYWG3fvl319fXO7dXV1dq3b59SUlLUu3dvRUVFqaSk\nxDh227ZtslgsSktL8/7kAACfcOvvGBYuXKjp06crJydHkydPVmNjowoLCyVJs2bNUlBQkHJzc7Vs\n2TKFhIRo2LBh2r9/v2w2m7KystStWzefngQAwHvcCkNaWpo2bNigF198UU8++aT8/PxktVq1evVq\ndenSRZL0yCOPKDAwUBs3blRJSYk6duyoadOmacaMGT49AQCAd7kVBkmyWq1av379t64ZP368xo8f\nf70zAQBaEO8jBQAYCAMAwOD2S0k3o6amJknSyZMnW3gSAGg9vvqZ+dXP0K9r1WE4ffq0JCk7O7uF\nJwGA1uf06dPONxD9N4vD4XC0wDxeUV9fL7vdrg4dOhh/lQ0A+GZNTU06ffq0EhMTFRIS0mx/qw4D\nAMD7uPkMADAQBgCAgTAAAAyEAQBgIAwAAANhAAAYCAMAwEAYAAAGwgAAMBAGAICBMAAADIQBcENj\nY6PWrFmjn/3sZ7r77rvVq1cvDR8+3Pivhg6HQ7/73e80dOhQ9erVS5mZmXr//fc1ZMgQLViwwLnu\n8uXLstlsysjIUGJiogYMGKD8/HzV19e3wJkBzbXqj90GbpRFixZp+/btmj17tu68807V1tbqtdde\nU15enrp06aKBAwcqLy9PGzdu1IQJE/TTn/5UBw8e1Pz585t95v3Pf/5zffDBB8rJyVFycrLKysr0\n0ksv6dChQyoqKpKfH7+voWURBsCFhoYGVVdXa968eZo0aZJze3Jysvr166ddu3ape/fu2rRpkyZO\nnKiFCxdKkgYMGKCOHTs6v5ek3bt367333tMTTzyhKVOmONfFxcVp5syZeu+993T//fff2BMEvoYw\nAC4EBQVp3bp1kqTz58/riy++UGVlpQ4cOCDpajh27dqlK1euKDMz0zh25MiReuaZZ5zf/+Uvf5Ek\nDR06VJcvX3ZuHzBggEJDQ7Vz507CgBZHGAA37N27VytWrNCnn36qwMBAxcfHq3fv3pKu3luorq6W\nJLVv3944LjAwUG3btnV+f+bMGUnSkCFDrvk8p06d8sX4gEcIA+BCZWWlpkyZoj59+qikpETx8fHy\n9/dXXV2dfv/730uSoqKiJF39V4k/+tGPnMdevnxZZ8+edX4fGRkpSSouLlZwcHCz5woPD/flqQBu\n4S4X4ILdbtfFixeVk5OjO+64w/lvZD/44ANJV68Y+vfvL4vFopKSEuPY7du3GzefU1NTJUn//ve/\n1atXL+dX27ZtlZ+fr88+++wGnRXwzbhiAFxITExUYGCgCgoK1NDQoODgYO3evVsbNmyQxWJRXV2d\nbrvtNmVnZ6u4uFh+fn669957VV5erjVr1kiSLBaLJKl///4aOHCgfvGLX6i8vFx33323vvzyS61Z\ns0Y1NTVKSkpqyVMFJPE/nwG3fPTRRyooKNCxY8cUHBysuLg4TZo0Sdu2bdORI0dUWloqh8Mhm82m\nLVu26Msvv1RcXJwef/xx5ebmauLEiVq0aJGkqzer165dqz/+8Y86ceKEoqKidM8992jWrFm64447\nWvhMAcIAeMWZM2f00UcfKTU1VZ06dXJuP3z4sEaNGqVnnnlG2dnZLTgh4D7CAHhBXV2dBg4cqM6d\nO2vq1Klq3769qqqqtHbtWjU0NOgPf/iD8wY1cLMjDICXlJWVyWazac+ePaqpqVHbtm2Vlpam2bNn\nKzo6uqXHA9xGGAAABt6uCgAwEAYAgIEwAAAMhAEAYCAMAAADYQAAGP4X9kUgKveMLDEAAAAASUVO\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a131aa890>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "box_hist('google','affiliate_provider')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaUAAAEMCAYAAABkwamIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a137d27d0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEMCAYAAADAqxFbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz\nAAALEgAACxIB0t1+/AAAADl0RVh0U29mdHdhcmUAbWF0cGxvdGxpYiB2ZXJzaW9uIDIuMS4wLCBo\ndHRwOi8vbWF0cGxvdGxpYi5vcmcvpW3flQAAFlJJREFUeJzt3X9U1fXhx/HX5fcvITVFzygBSW1i\noV1kkkPMH5jGVDyZiRontaOiptN+LF2d1ESm6cYKby7DX9TqpKdN1uxrVLMzd5zmSvEHHmYkzGky\nRXGIIN7vH367371DvffOe0H0+TiHP3h/3p/L+/MPTz4/uNdit9vtAgDg//i09AIAADcXwgAAMBAG\nAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAMhAEAYCAMAAADYQAAGPxa\negE3oq6uTiUlJerQoYN8fX1bejkA0Co0Njbq1KlTio+PV1BQUJPtrToMJSUlyszMbOllAECrVFhY\nKKvV2mS8VYehQ4cOkq4cXKdOnVp4NQDQOpw4cUKZmZmO36Hf16rD8N3lo06dOikqKqqFVwMArcu1\nLsFz8xkAYHA5DDt37tTjjz+uBx54QP3799e8efP0z3/+05jzpz/9SY8++qgSEhL04IMPavHixTp/\n/rzHFw0A8B6XwrB7925NmTJFfn5+Wr58uebPn6+//e1vmjhxompqaiRJO3bs0PTp0xUZGamVK1dq\n8uTJ2rx5s2bOnOnVAwAAeJZL9xjeffddBQUF6Y033lBISIgkKTIyUllZWSouLtaoUaO0fPly/fCH\nP9Svf/1rWSwWSVLnzp01Z84c7dixQykpKd47CgCAx7h0xnDx4kX5+/srODjYMda+fXtJUnV1tY4f\nP64jR47o4YcfdkRBkgYPHqyQkBAVFxd7eNkAAG9xKQwTJkxQbW2tcnNzdfr0aVVUVGjp0qWKiIhQ\nWlqaysrKJEldu3Y19vP391dUVJRjOwDg5ufSpaSkpCTNmTNHK1asUEFBgSSpTZs2WrNmjTp37qy9\ne/dKksLCwprsGxoa6rgP4Y6MjAync+rr691+XQDA9bkUhsWLF2vTpk167LHHNGzYMF24cEFvv/22\nJk+erNdff12XL1+WJOMy0n/y8eGpWABoLZyG4eTJkyosLNSoUaO0aNEix3hqaqoee+wxLVy4UC+9\n9JIkXfXR1H//+9+644473F7Yli1bnM6prKzUoEGD3H5tAMC1Of1T/vjx47Lb7UpMTDTGfX19ZbVa\n9Y9//EORkZGSpK+//tqY09DQoIqKCsXFxXlwyQAAb3IahujoaPn5+WnXrl3GeGNjo/bs2aP27dur\ne/fuiouL0x//+EfHZSVJ+vjjj3XhwgUNGDDA8ysHAHiF00tJbdu21VNPPaX8/Hz5+flp2LBhunjx\not59913t379fy5Ytk8Vi0dy5c5Wdna0ZM2bo0Ucf1bFjx/SrX/1K/fr1U2pqajMcCgDAE1y6+fz0\n008rKipKmzZtUlFRkUJDQ9WjRw9t3LhRffv2lXTlfxby8vKUn5+vOXPm6I477tCYMWM0d+5crx4A\nAMCzXH531TFjxmjMmDHXnZOWlqa0tLQbXhQAoOXwHCkAwEAYAAAGwgAAMBAGAICBMAAADIQBAGAg\nDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQ\nBgCAgTAAAAyEAQBgIAwAAANhAFwUHx8vi8Xi1a/4+PiWPkyAMACuKikpkd1ud/mry3NFbs232+0q\nKSlp6cMECAMAwEQYAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYA\ngMHP2YTKykoNGjTounNKS0tVXV2tpKSkq24vKChQcnLyf7dCAECzchqGjh07qrCwsMn4/v37tWzZ\nMmVmZkqSDh48KElasmSJYmJijLndu3f3xFoBAM3AaRgCAgJktVqNserqas2fP19Wq1UvvPCCpCtv\nSezr66tHHnlEwcHB3lktAMDrnIbhalasWKHTp09r3bp18vO78hIHDx5UbGwsUQCAVs7tMJSUlOj9\n99/XzJkzFR0d7Rg/cOCAwsLCNGXKFO3du1eXL1/Wj370Iz3zzDPq2rWr2wvLyMhwOqe+vt7t1wUA\nXJ/bTyXZbDZFREQoKyvLMVZTU6OKigqVl5crJSVFa9as0YIFC1RaWqpx48bp2LFjnlwzAMCL3Dpj\nqKioUHFxsWbMmKGwsDDHeEBAgAoKCtShQwfFxcVJkqxWq6xWq9LT07V69Wrl5OS4tbAtW7Y4nePK\nE1MAAPe4FYYPP/xQdrtdo0ePNsYDAwPVr1+/JvNjYmIUGxurQ4cO3dgqAQDNxq1LSdu3b1fv3r0V\nFRVljJeXl6uwsFAnTpxosk9dXZ3atWt3Y6sEADQbl8NQV1enQ4cOKTExscm2s2fPatGiRXrnnXeM\n8S+++ELHjh276tkEAODm5PKlpCNHjujSpUvq1q1bk2333XefBg8erDfffFOSlJiYqLKyMuXn5ysu\nLk6TJk3y3IoBAF7lchiqqqokSeHh4U22WSwWvfrqq1q7dq22bt2qt956SxERERoxYoRmz56twMBA\nz60YAOBVLofhoYceUmlp6TW3BwUFKTs7W9nZ2R5ZGACgZfDuqgAAA2EAABgIAwDAQBgAAAbCAAAw\nEAYAgIEwAAAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBgIAwAAANhAAAY\nCAMAwEAYAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAM\nhAEAYCAMAAADYQAAGAgDAMBAGAAABj9nEyorKzVo0KDrziktLZUkffDBB1q7dq2++eYbtWvXTqNG\njdKMGTMUEBDgmdUCALzOaRg6duyowsLCJuP79+/XsmXLlJmZKUl677339POf/1xjx47VvHnz9NVX\nX+mNN97QqVOn9Morr3h+5QAAr3AahoCAAFmtVmOsurpa8+fPl9Vq1QsvvKCLFy9q5cqVSktL0+LF\niyVJqampCg0N1YoVK/Tkk0+qa9eu3jkCAIBH/Vf3GFasWKHTp0/rlVdekZ+fn/bt26czZ85o+PDh\nxrz09HTZ7XYVFxd7ZLEAAO9zOwwlJSV6//339dRTTyk6OlqSVFZWJklNzgoiIyMVEhLi2A4AuPk5\nvZT0fTabTREREcrKynKM1dTUSJLatGnTZH5YWJhjuzsyMjKczqmvr3f7dQEA1+fWGUNFRYWKi4s1\nYcIEhYWFOcYvX74sSbJYLFf/IT48FQsArYVbZwwffvih7Ha7Ro8ebYyHh4dLunLmEBkZaWw7f/78\nVc8knNmyZYvTOa48SgsAcI9bf8pv375dvXv3VlRUlDH+3b2F8vJyY/zkyZOqra1VXFzcja0SANBs\nXA5DXV2dDh06pMTExCbbevfurYiICBUVFRnjW7dulcViUUpKyo2vFADQLFy+lHTkyBFdunRJ3bp1\na7ItICBA2dnZWrp0qYKCgjRs2DDt27dPNptNGRkZV90HAHBzcjkMVVVVkv7/fsL3PfHEE/L399eG\nDRtUVFSkjh07atq0aZoxY4ZnVgoAaBYuh+Ghhx5yvCfStYwfP17jx4+/4UUBAFoOz5ECAAyEAQBg\nIAwAAANhAAAYCAMAwEAYAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDA4NZHewK3ivtf\n/h+dvdDg9Z8T/fwfvP4zIoL99dVLQ73+c3D7IAy4LZ290KDyZSNaehke0Rzxwe2FS0kAAANhAAAY\nCAMAwEAYAAAGwgAAMBAGAICBMAAADIQBAGAgDAAAA2EAABgIAwDAQBgAAAbCAAAwEAYAgIEwAAAM\nhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADD4uTqxrKxMq1at0l//+lc1Njaqe/fumj17tvr16ydJ\nqq6uVlJS0lX3LSgoUHJysmdWDADwKpfCcPToUY0bN0533XWXFi1aJH9/f61bt05Tp07Vpk2blJCQ\noIMHD0qSlixZopiYGGP/7t27e37lAACvcCkMubm5Cg8P16ZNmxQaGipJSk5O1siRI/Xpp58qISFB\nJSUl8vX11SOPPKLg4GCvLhoA4D1Ow3Du3Dl9/vnnmjNnjiMKkhQSEqLt27c7vj948KBiY2OJAgC0\nck5vPh8+fFiNjY2KiorSkiVLlJycrJ49e2rs2LHas2ePY96BAwfk7++vKVOmqE+fPkpISNC0adP0\n97//3asHAADwLKdnDFVVVZKknJwc9ejRQzk5OWpoaJDNZlNWVpY2btyouLg4VVRUKDg4WKNHj9a0\nadP09ddfKz8/X+PGjdPmzZt19913u7WwjIwMp3Pq6+vdek0AgHNOw9DQ0CBJuvPOO2Wz2eTr6ytJ\n6tu3r4YOHaq8vDzZbDYVFBSoQ4cOiouLkyRZrVZZrValp6dr9erVysnJ8eJhAAA8xWkYwsLCJEmp\nqamOKEhSeHi4+vTpo927dyswMNDx2Op/iomJUWxsrA4dOuT2wrZs2eJ0TmVlpQYNGuT2awMArs3p\nPYbvHj29ePFik2319fUKCgpSeXm5CgsLdeLEiSZz6urq1K5dOw8sFQDQHJyGITY2VtHR0dq2bZvq\n6uoc41VVVdq7d6+SkpJ09uxZLVq0SO+8846x7xdffKFjx45d9WwCAHBzcun/GBYsWKDp06crKytL\nkydPVkNDg/Lz8yVJs2bN0t13363BgwfrzTfflCQlJiaqrKxM+fn5iouL06RJk7x3BAAAj3IpDCkp\nKVq/fr1ee+01Pfvss/Lx8ZHVatWqVavUpUsXSdKrr76qtWvXauvWrXrrrbcUERGhESNGaPbs2QoM\nDPTqQQAAPMfl90qyWq1at27dNbcHBQUpOztb2dnZnlgXAKCF8O6qAAADYQAAGAgDAMDg8j0G4FYS\nErNKvdY/39LL8IiQmEhJI1p6GbiFEAbclmq/nqvyZbfGL9Po5//Q0kvALYZLSQAAA2EAABgIAwDA\nQBgAAAbCAAAwEAYAgIEwAAAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBg\nIAwAAANhAAAY+GhP3LZulY/EjAj2b+kl4BZDGHBbao7Pe45+/g+3zOdK4/bCpSQAgIEwAAAMhAEA\nYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBgIAwAAIPLYSgrK1N2drYSExPVp08f\nPf744/rLX/5izPnggw+Unp6u++67T6mpqfrlL3+p+vp6jy8aAOA9LoXh6NGjGjdunI4fP65Fixbp\nF7/4hXx9fTV16lR9+eWXkqT33ntPzz33nBISEpSXl6fRo0drzZo1evnll716AAAAz3Lpbbdzc3MV\nHh6uTZs2KTQ0VJKUnJyskSNH6tNPP9W9996rlStXKi0tTYsXL5YkpaamKjQ0VCtWrNCTTz6prl27\neu8oAAAe4/SM4dy5c/r88881btw4RxQkKSQkRNu3b9fcuXO1b98+nTlzRsOHDzf2TU9Pl91uV3Fx\nsedXDgDwCqdhOHz4sBobGxUVFaUlS5YoOTlZPXv21NixY7Vnzx5JV+4/SGpyVhAZGamQkBDHdgDA\nzc/ppaSqqipJUk5Ojnr06KGcnBw1NDTIZrMpKytLGzduVE1NjSSpTZs2TfYPCwtzbHdHRkaG0znc\n2AYAz3MahoaGBknSnXfeKZvNJl9fX0lS3759NXToUOXl5SkpKUmSZLFYrvoaPj48FYvWLz4+XgcO\nHHBrH0uuez+jZ8+eKikpcW8nwMOchiEsLEzSlZvJ30VBksLDw9WnTx/t3r1bQ4YMkSTV1NQoMjLS\n2P/8+fNXPZNwZsuWLU7nVFZWatCgQW6/NvDf4Bc2bhdO/5SPiYmRJF28eLHJtvr6egUFBTnuLZSX\nlxvbT548qdraWsXFxXlgqQCA5uA0DLGxsYqOjta2bdtUV1fnGK+qqtLevXuVlJSk3r17KyIiQkVF\nRca+W7dulcViUUpKiudXDgDwCpf+j2HBggWaPn26srKyNHnyZDU0NCg/P1+SNGvWLAUEBCg7O1tL\nly5VUFCQhg0bpn379slmsykjI0PdunXz6kEAADzHpTCkpKRo/fr1eu211/Tss8/Kx8dHVqtVq1at\nUpcuXSRJTzzxhPz9/bVhwwYVFRWpY8eOmjZtmmbMmOHVAwAAeJZLYZAkq9WqdevWXXfO+PHjNX78\n+BtdEwCgBfEcKQDAQBgAAAaXLyXdjBobGyVJJ06caOGVAEDr8d3vzO9+h35fqw7DqVOnJEmZmZkt\nvBIAaH1OnTrleIDoP1nsdru9BdbjEXV1dSopKVGHDh2M/8oGAFxbY2OjTp06pfj4eAUFBTXZ3qrD\nAADwPG4+AwAMhAEAYCAMAAADYQAAGAgDAMBAGAAABsIAADAQBgCAgTAAAAyEAQBgIAwAAANhAFzQ\n0NCg1atX6yc/+Ynuv/9+9erVS8OHDzc+1dBut+s3v/mNhg4dql69eik9PV0ff/yxhgwZoueff94x\n79KlS7LZbEpLS1N8fLwGDBig3Nxc1dXVtcCRAU216rfdBprLwoULtW3bNs2ePVv33nuvampq9Pbb\nbysnJ0ddunTRwIEDlZOTow0bNmjChAn68Y9/rAMHDmj+/PlN3vP+pz/9qT755BNlZWUpMTFRpaWl\nev3113Xw4EEVFBTIx4e/19CyCAPgRH19vaqqqjRv3jxNmjTJMZ6YmKh+/fpp586d6t69uzZu3KiJ\nEydqwYIFkqQBAwaoY8eOju8ladeuXfroo4/0zDPPaMqUKY55MTExmjlzpj766CM9/PDDzXuAwPcQ\nBsCJgIAArV27VpJ07tw5ffPNN6qoqND+/fslXQnHzp07dfnyZaWnpxv7jhw5Ui+++KLj+z//+c+S\npKFDh+rSpUuO8QEDBig4OFg7duwgDGhxhAFwwZ49e7R8+XJ9+eWX8vf3V2xsrHr37i3pyr2Fqqoq\nSVL79u2N/fz9/dW2bVvH96dPn5YkDRky5Ko/5+TJk95YPuAWwgA4UVFRoSlTpqhPnz4qKipSbGys\nfH19VVtbq9/+9reSpIiICElXPirxBz/4gWPfS5cu6cyZM47vw8PDJUmFhYUKDAxs8rNCQ0O9eSiA\nS7jLBThRUlKiCxcuKCsrS/fcc4/jY2Q/+eQTSVfOGPr37y+LxaKioiJj323bthk3n5OTkyVJ//rX\nv9SrVy/HV9u2bZWbm6uvvvqqmY4KuDbOGAAn4uPj5e/vr7y8PNXX1yswMFC7du3S+vXrZbFYVFtb\nq7vuukuZmZkqLCyUj4+PHnzwQZWVlWn16tWSJIvFIknq37+/Bg4cqJ/97GcqKyvT/fffr2+//Var\nV69WdXW1EhISWvJQAUl85jPgks8++0x5eXk6evSoAgMDFRMTo0mTJmnr1q06fPiwiouLZbfbZbPZ\ntHnzZn377beKiYnR008/rezsbE2cOFELFy6UdOVm9Zo1a/T73/9ex48fV0REhB544AHNmjVL99xz\nTwsfKUAYAI84ffq0PvvsMyUnJ6tTp06O8UOHDmnUqFF68cUXlZmZ2YIrBFxHGAAPqK2t1cCBA9W5\nc2dNnTpV7du3V2VlpdasWaP6+nr97ne/c9ygBm52hAHwkNLSUtlsNu3evVvV1dVq27atUlJSNHv2\nbEVGRrb08gCXEQYAgIHHVQEABsIAADAQBgCAgTAAAAyEAQBgIAwAAMP/AhtKIC4JnV0jAAAAAElF\nTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a13843d10>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "box_hist('direct','affiliate_provider')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "We had a research on the old group people as the young age group has the similar distribution in both 'direct' and 'google'.\n",
    "From the box plots and histogram plots we can see the user using google has a larger variance and more popular in the old group whose age is larger than 60. This means they may reach the google to get to know ther airbnb first."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Age composition in omg group"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaUAAAEMCAYAAABkwamIAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\noURERJLBUCIiIsn4Pyt6/AFKsS1UAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a137d2a50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYYAAAEMCAYAAADAqxFbAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nRK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a14304f90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "box_hist('omg','first_affiliate_tracked')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "From this part, we can draw the conclusion that the variance of 'OMG' group is much larger than the average group. This means this group may have a wider customer range."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "In  this part, we want to figure out how users interacted with \"Sharing Economy\" by looking into the distribution of assorted records;\n",
    "\n",
    "In the figure of # Visulization of affiliate_Provider, we can see that the 'direct' method outnumbers other methods in a siginificant scale,\n",
    "This indicates that most users reach the airbnb directly. This means that they mght have known and accepted the concepts of \"sharing economy\" and willing to give it a shot. To further boraden the market, we may want to spread the sharing economy first and then the user may want to accept the concept.\n",
    "\n",
    "\n",
    "The next thing we focus on is the old age group between the 'google' and 'direct' way, from this picture we can see that though the number difference between the 'direct' and 'google' is relatively large, the difference in the old group is not obvious, this support our conslusion that the airbnb is not that popular in the old group (judged by whether they can reach airbnb by 'direct' way). The reason for this result may be that the sharing economy is not that popular in the old group. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Sessions Part"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Extract user_id and destination for easier merge, shorten_users contains id and destination only"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>country_destination</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gxn3p5htnn</td>\n",
       "      <td>NDF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>820tgsjxq7</td>\n",
       "      <td>NDF</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4ft3gnwmtx</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bjjt8pjhuk</td>\n",
       "      <td>other</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>87mebub9p4</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id country_destination\n",
       "0  gxn3p5htnn                 NDF\n",
       "1  820tgsjxq7                 NDF\n",
       "2  4ft3gnwmtx                  US\n",
       "3  bjjt8pjhuk               other\n",
       "4  87mebub9p4                  US"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shorten_users = train_users[['id','country_destination']]\n",
    "shorten_users.rename(columns={'id': 'user_id'}, inplace=True) \n",
    "shorten_users.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize the distribution of destination in the training set"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "   country_destination  user_id\n",
      "0                   AU      539\n",
      "1                   CA     1428\n",
      "2                   DE     1061\n",
      "3                   ES     2249\n",
      "4                   FR     5023\n",
      "5                   GB     2324\n",
      "6                   IT     2835\n",
      "7                  NDF   124543\n",
      "8                   NL      762\n",
      "9                   PT      217\n",
      "10                  US    62376\n",
      "11               other    10094\n"
     ]
    }
   ],
   "source": [
    "print(shorten_users.groupby(['country_destination'])['user_id'].nunique().reset_index())"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Add labels to data, count = 1 means booking, count = 0 means NDF"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>country_destination</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>gxn3p5htnn</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>820tgsjxq7</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>4ft3gnwmtx</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>bjjt8pjhuk</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>87mebub9p4</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id country_destination count\n",
       "0  gxn3p5htnn                 NDF     0\n",
       "1  820tgsjxq7                 NDF     0\n",
       "2  4ft3gnwmtx                  US     1\n",
       "3  bjjt8pjhuk               other     1\n",
       "4  87mebub9p4                  US     1"
      ]
     },
     "execution_count": 23,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "shorten_users['count']=shorten_users['country_destination']\n",
    "shorten_users.loc[shorten_users['count'] != 'NDF','count'] = 1\n",
    "shorten_users.loc[shorten_users['count'] == 'NDF','count'] = 0\n",
    "\n",
    "shorten_users.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Merge the info in sessions and training data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>country_destination</th>\n",
       "      <th>count</th>\n",
       "      <th>action</th>\n",
       "      <th>action_type</th>\n",
       "      <th>action_detail</th>\n",
       "      <th>device_type</th>\n",
       "      <th>secs_elapsed</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>d1mm9tcy42</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>lookup</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>319.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>d1mm9tcy42</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>search_results</td>\n",
       "      <td>click</td>\n",
       "      <td>view_search_results</td>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>67753.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>d1mm9tcy42</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>lookup</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>301.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>d1mm9tcy42</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>search_results</td>\n",
       "      <td>click</td>\n",
       "      <td>view_search_results</td>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>22141.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>d1mm9tcy42</td>\n",
       "      <td>other</td>\n",
       "      <td>1</td>\n",
       "      <td>lookup</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>435.0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id country_destination count          action action_type  \\\n",
       "0  d1mm9tcy42               other     1          lookup         NaN   \n",
       "1  d1mm9tcy42               other     1  search_results       click   \n",
       "2  d1mm9tcy42               other     1          lookup         NaN   \n",
       "3  d1mm9tcy42               other     1  search_results       click   \n",
       "4  d1mm9tcy42               other     1          lookup         NaN   \n",
       "\n",
       "         action_detail      device_type  secs_elapsed  \n",
       "0                  NaN  Windows Desktop         319.0  \n",
       "1  view_search_results  Windows Desktop       67753.0  \n",
       "2                  NaN  Windows Desktop         301.0  \n",
       "3  view_search_results  Windows Desktop       22141.0  \n",
       "4                  NaN  Windows Desktop         435.0  "
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sessions = pd.merge(shorten_users,sessions,on='user_id',how='inner')\n",
    "sessions.head()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Visualize the variaty of each keywords in the merged data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "user_id                 73815\n",
       "country_destination        12\n",
       "count                       2\n",
       "action                    331\n",
       "action_type                 9\n",
       "action_detail             128\n",
       "device_type                14\n",
       "secs_elapsed           256593\n",
       "dtype: int64"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sessions.apply(lambda x: x.nunique(),axis=0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Time Interval Analysis (Time Interval is the time one user spends on Airbnb)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Count the total time of each user"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>secs_elapsed</th>\n",
       "      <th>country_destination</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00023iyk9l</td>\n",
       "      <td>867896.0</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>001wyh0pz8</td>\n",
       "      <td>282965.0</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0028jgx1x1</td>\n",
       "      <td>297010.0</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>002qnbzfs5</td>\n",
       "      <td>6487080.0</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0035hobuyj</td>\n",
       "      <td>5724670.0</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id  secs_elapsed country_destination count\n",
       "0  00023iyk9l      867896.0                  US     1\n",
       "1  001wyh0pz8      282965.0                 NDF     0\n",
       "2  0028jgx1x1      297010.0                 NDF     0\n",
       "3  002qnbzfs5     6487080.0                  US     1\n",
       "4  0035hobuyj     5724670.0                  US     1"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "grpby = sessions.groupby(['user_id'])['secs_elapsed'].sum().reset_index()\n",
    "grpby.columns = ['user_id','secs_elapsed']\n",
    "grpby = pd.merge(grpby,shorten_users,on='user_id',how='inner')\n",
    "grpby.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "a0972945f948411ebcf0d100f4c5dfb6",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.timeIntervalVisual>"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from ipywidgets import interact, interactive, fixed, interact_manual\n",
    "import ipywidgets as widgets\n",
    "\n",
    "def timeIntervalVisual(groupNum, typeCurr):\n",
    "    '''\n",
    "    This is the visualization function for time interval distribution\n",
    "    *param: groupNum is the number of groups you want to divided the time intervals into\n",
    "    '''\n",
    "    assert groupNum > 1\n",
    "    types = ['All', 'Book', 'NDF']\n",
    "    assert typeCurr in types\n",
    "    grpby['timeclass'] = pd.qcut(grpby['secs_elapsed'], groupNum)\n",
    "    time_raw = grpby\n",
    "    if typeCurr == types[1]:\n",
    "        time_raw = grpby[grpby['count'] == 1]\n",
    "    elif typeCurr == types[2]:\n",
    "        time_raw = grpby[grpby['count'] == 0]\n",
    "    times = time_raw[['timeclass', 'count']].groupby(['timeclass']).count().reset_index()\n",
    "    times.index = times['timeclass']\n",
    "    times.plot('timeclass', 'count', kind='pie', autopct='%1.1f%%', title = 'Time Interval(secs) Distribution', figsize = (10, 10))\n",
    "    plt.legend(loc='center left', bbox_to_anchor=(1.3, 0.5))\n",
    "    plt.axis('off')\n",
    "    plt.show()\n",
    "    \n",
    "interact(timeIntervalVisual, groupNum = widgets.IntSlider(min = 2, max = 10, step = 1, value = 5), \\\n",
    "        typeCurr = widgets.Dropdown(options = {'All', 'Book', 'NDF'}))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Conclusion: We can see that the Booked users spent more time on the Airbnb"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Action Type Analysis (Action Type is the action one user is recorded in his activity log)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Firstly, see the distribution of user action type on action level"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a12b06190>"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAbYAAAGXCAYAAADf+XXjAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nW/8K0fZTYgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x10ea8ff90>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sessions.action_type.value_counts().plot(kind= 'bar', title = 'Action Type Distribution')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Remove the unknown one"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {
    "collapsed": true
   },
   "outputs": [],
   "source": [
    "infos = []\n",
    "for key in sessions.action_type.unique()[1:]:\n",
    "    if key != np.nan and key != '-unknown-':\n",
    "        total = len(sessions[sessions.action_type == key])\n",
    "        if total != 0:\n",
    "            book = sessions[sessions.action_type == key]['count'].sum()\n",
    "            percent = 1.0 * book / total * 100\n",
    "            infos.append([key, book, total, percent])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Percentage of each action_type regarding booking class and total numbers, action level statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>action_type</th>\n",
       "      <th>book #</th>\n",
       "      <th>total #</th>\n",
       "      <th>booking percent</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>booking_request</td>\n",
       "      <td>9081</td>\n",
       "      <td>10695</td>\n",
       "      <td>84.9088359046</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>message_post</td>\n",
       "      <td>32299</td>\n",
       "      <td>55567</td>\n",
       "      <td>58.1262259974</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>partner_callback</td>\n",
       "      <td>3798</td>\n",
       "      <td>7005</td>\n",
       "      <td>54.2184154176</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>booking_response</td>\n",
       "      <td>1</td>\n",
       "      <td>2</td>\n",
       "      <td>50.0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>click</td>\n",
       "      <td>481757</td>\n",
       "      <td>1064436</td>\n",
       "      <td>45.2593674021</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>view</td>\n",
       "      <td>748404</td>\n",
       "      <td>1669723</td>\n",
       "      <td>44.8220453333</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>data</td>\n",
       "      <td>516182</td>\n",
       "      <td>1188495</td>\n",
       "      <td>43.4315668135</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>submit</td>\n",
       "      <td>138407</td>\n",
       "      <td>357371</td>\n",
       "      <td>38.7292197744</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "        action_type  book #  total # booking percent\n",
       "5   booking_request    9081    10695   84.9088359046\n",
       "4      message_post   32299    55567   58.1262259974\n",
       "6  partner_callback    3798     7005   54.2184154176\n",
       "7  booking_response       1        2            50.0\n",
       "0             click  481757  1064436   45.2593674021\n",
       "2              view  748404  1669723   44.8220453333\n",
       "1              data  516182  1188495   43.4315668135\n",
       "3            submit  138407   357371   38.7292197744"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "action_Summary = pd.DataFrame(np.array(infos), columns= ['action_type', 'book #', 'total #', 'booking percent'])\n",
    "action_Summary.sort_values('booking percent', ascending = 0)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "distribution of each action type on user level"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a13147a10>"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAaMAAAGjCAYAAACBlXr0AAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nu/8DM/qRtc6Gty4AAAAASUVORK5CYII=\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a13c9cfd0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sessions.groupby(['action_type'])['user_id'].nunique().reset_index().plot('action_type', 'user_id', kind = 'bar')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>-unknown-</th>\n",
       "      <th>booking_request</th>\n",
       "      <th>booking_response</th>\n",
       "      <th>click</th>\n",
       "      <th>data</th>\n",
       "      <th>message_post</th>\n",
       "      <th>partner_callback</th>\n",
       "      <th>submit</th>\n",
       "      <th>view</th>\n",
       "      <th>country_destination</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00023iyk9l</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>9</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>21</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>001wyh0pz8</td>\n",
       "      <td>6</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>66</td>\n",
       "      <td>2</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>8</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0028jgx1x1</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>9</td>\n",
       "      <td>5</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>002qnbzfs5</td>\n",
       "      <td>184</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>140</td>\n",
       "      <td>140</td>\n",
       "      <td>16</td>\n",
       "      <td>0</td>\n",
       "      <td>15</td>\n",
       "      <td>216</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0035hobuyj</td>\n",
       "      <td>9</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>206</td>\n",
       "      <td>41</td>\n",
       "      <td>3</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>55</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id  -unknown-  booking_request  booking_response  click  data  \\\n",
       "0  00023iyk9l          0                1                 0      4     9   \n",
       "1  001wyh0pz8          6                0                 0     66     2   \n",
       "2  0028jgx1x1          1                0                 0      9     5   \n",
       "3  002qnbzfs5        184                1                 0    140   140   \n",
       "4  0035hobuyj          9                0                 0    206    41   \n",
       "\n",
       "   message_post  partner_callback  submit  view country_destination count  \n",
       "0             1                 1       0    21                  US     1  \n",
       "1             0                 0       3     8                 NDF     0  \n",
       "2             0                 0       1    15                 NDF     0  \n",
       "3            16                 0      15   216                  US     1  \n",
       "4             3                 0       4    55                  US     1  "
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "action_type = pd.pivot_table(sessions, index = ['user_id'],columns = ['action_type'],\n",
    "                             values = 'action',aggfunc=len,fill_value=0).reset_index()\n",
    "action_type= pd.merge(action_type,shorten_users,on='user_id',how='inner')\n",
    "action_type.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "7ce2c6ba1e364d04b6ffa02679b5d57f",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.actionTypeVisual>"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def actionTypeVisual(typeCurr):\n",
    "    '''\n",
    "    This function performs the visualization of the action type distribution\n",
    "    *param typeCurr: the group of data you want to show visualization with\n",
    "            \"All\" means all the data, \"Book\" means only the booked users and \"NDF\" means the not booked ones\n",
    "    '''\n",
    "    types = ['All', 'Book', 'NDF']\n",
    "    assert typeCurr in types\n",
    "    type_info = [['all', 0]]\n",
    "    action_raw = action_type\n",
    "    if typeCurr == types[1]:\n",
    "        action_raw = action_type[action_type['count'] == 1]\n",
    "    elif typeCurr == types[2]:\n",
    "        action_raw = action_type[action_type['count'] == 0]\n",
    "    for k in action_raw.columns[2:-2]:\n",
    "        count = action_raw[k].sum()\n",
    "        type_info.append([k, count])\n",
    "        type_info[0][1] += int(count)\n",
    "    type_info = pd.DataFrame(type_info, columns=['action_type', 'count'])\n",
    "    type_info.plot('action_type', 'count', kind='bar', title = typeCurr, color='#FD5C64', rot=90)\n",
    "interact(actionTypeVisual, typeCurr = widgets.Dropdown(options = {'All', 'Book', 'NDF'}))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Device Type Analysis (Device Type is the device that one user used to perform his action)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>device_type</th>\n",
       "      <th>user_id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>-unknown-</td>\n",
       "      <td>13126</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Android App Unknown Phone/Tablet</td>\n",
       "      <td>2757</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Android Phone</td>\n",
       "      <td>3474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Blackberry</td>\n",
       "      <td>59</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Chromebook</td>\n",
       "      <td>260</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Linux Desktop</td>\n",
       "      <td>354</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Mac Desktop</td>\n",
       "      <td>28893</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Opera Phone</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Tablet</td>\n",
       "      <td>848</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Windows Desktop</td>\n",
       "      <td>24768</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>10</th>\n",
       "      <td>Windows Phone</td>\n",
       "      <td>50</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>iPad Tablet</td>\n",
       "      <td>6283</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>12</th>\n",
       "      <td>iPhone</td>\n",
       "      <td>16054</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>iPodtouch</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                         device_type  user_id\n",
       "0                          -unknown-    13126\n",
       "1   Android App Unknown Phone/Tablet     2757\n",
       "2                      Android Phone     3474\n",
       "3                         Blackberry       59\n",
       "4                         Chromebook      260\n",
       "5                      Linux Desktop      354\n",
       "6                        Mac Desktop    28893\n",
       "7                        Opera Phone        1\n",
       "8                             Tablet      848\n",
       "9                    Windows Desktop    24768\n",
       "10                     Windows Phone       50\n",
       "11                       iPad Tablet     6283\n",
       "12                            iPhone    16054\n",
       "13                         iPodtouch       41"
      ]
     },
     "execution_count": 34,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sessions.groupby(['device_type'])['user_id'].nunique().reset_index()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "people may have more than one devices"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>device_type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00023iyk9l</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>001wyh0pz8</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0028jgx1x1</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>002qnbzfs5</td>\n",
       "      <td>2</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0035hobuyj</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id  device_type\n",
       "0  00023iyk9l            2\n",
       "1  001wyh0pz8            1\n",
       "2  0028jgx1x1            2\n",
       "3  002qnbzfs5            2\n",
       "4  0035hobuyj            1"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sessions.groupby(['user_id'])['device_type'].nunique().reset_index().head() "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Device"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>user_id</th>\n",
       "      <th>-unknown-</th>\n",
       "      <th>Android App Unknown Phone/Tablet</th>\n",
       "      <th>Android Phone</th>\n",
       "      <th>Blackberry</th>\n",
       "      <th>Chromebook</th>\n",
       "      <th>Linux Desktop</th>\n",
       "      <th>Mac Desktop</th>\n",
       "      <th>Opera Phone</th>\n",
       "      <th>Tablet</th>\n",
       "      <th>Windows Desktop</th>\n",
       "      <th>Windows Phone</th>\n",
       "      <th>iPad Tablet</th>\n",
       "      <th>iPhone</th>\n",
       "      <th>iPodtouch</th>\n",
       "      <th>country_destination</th>\n",
       "      <th>count</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>00023iyk9l</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>001wyh0pz8</td>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>0028jgx1x1</td>\n",
       "      <td>30</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>NDF</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>002qnbzfs5</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>775</td>\n",
       "      <td>0</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>0035hobuyj</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>489</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>US</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "      user_id  -unknown-  Android App Unknown Phone/Tablet  Android Phone  \\\n",
       "0  00023iyk9l          0                                 0              0   \n",
       "1  001wyh0pz8          0                                90              0   \n",
       "2  0028jgx1x1         30                                 0              1   \n",
       "3  002qnbzfs5         14                                 0              0   \n",
       "4  0035hobuyj          0                                 0              0   \n",
       "\n",
       "   Blackberry  Chromebook  Linux Desktop  Mac Desktop  Opera Phone  Tablet  \\\n",
       "0           0           0              0           36            0       0   \n",
       "1           0           0              0            0            0       0   \n",
       "2           0           0              0            0            0       0   \n",
       "3           0           0              0            0            0       0   \n",
       "4           0           0              0          489            0       0   \n",
       "\n",
       "   Windows Desktop  Windows Phone  iPad Tablet  iPhone  iPodtouch  \\\n",
       "0                0              0            0       4          0   \n",
       "1                0              0            0       0          0   \n",
       "2                0              0            0       0          0   \n",
       "3                0              0            0     775          0   \n",
       "4                0              0            0       0          0   \n",
       "\n",
       "  country_destination count  \n",
       "0                  US     1  \n",
       "1                 NDF     0  \n",
       "2                 NDF     0  \n",
       "3                  US     1  \n",
       "4                  US     1  "
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "device_type = pd.pivot_table(sessions, index = ['user_id'],columns = ['device_type'],values = 'action',\n",
    "                             aggfunc=len,fill_value=0).reset_index()\n",
    "device_type= pd.merge(device_type,shorten_users,on='user_id',how='inner')\n",
    "device_type.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "2ea67e009e5d4389a74fc449a56e5f29",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.deviceTypeVisual>"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def deviceTypeVisual(typeCurr):\n",
    "    '''\n",
    "    This function performs the visualization of the device type distribution\n",
    "    *param typeCurr: the group of data you want to show visualization with\n",
    "            \"All\" means all the data, \"Book\" means only the booked users and \"NDF\" means the not booked ones\n",
    "    '''\n",
    "    types = ['All', 'Book', 'NDF']\n",
    "    assert typeCurr in types\n",
    "    type_info = [['all', 0]]\n",
    "    device_raw = device_type\n",
    "    if typeCurr == types[1]:\n",
    "        device_raw = device_type[device_type['count'] == 1]\n",
    "    elif typeCurr == types[2]:\n",
    "        device_raw = device_type[device_type['count'] == 0]\n",
    "    for k in device_raw.columns[2:-2]:\n",
    "        count = device_raw[k].sum()\n",
    "        type_info.append([k, count])\n",
    "        type_info[0][1] += int(count)\n",
    "    type_info = pd.DataFrame(type_info, columns=['device_type', 'count'])\n",
    "    type_info.plot('device_type', 'count', kind='bar', title = typeCurr, color='#FD5C64', rot=90)\n",
    "interact(deviceTypeVisual, typeCurr = widgets.Dropdown(options = {'All', 'Book', 'NDF'}))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Combine action type and time "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "f367f9d5763d4b6b92889ddd6f3843ff",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.actionTypeTime>"
      ]
     },
     "execution_count": 38,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "def actionTypeTime(typeCurr):\n",
    "    '''\n",
    "    This function performs the visualization of the total time elapsed within each device type\n",
    "    *param typeCurr: the group of data you want to show visualization with\n",
    "            \"All\" means all the data, \"Book\" means only the booked users and \"NDF\" means the not booked ones\n",
    "    '''\n",
    "    types = ['All', 'Book', 'NDF']\n",
    "    assert typeCurr in types\n",
    "    action_time = sessions\n",
    "    if typeCurr == types[1]:\n",
    "        action_time = sessions[sessions['count'] == 1]\n",
    "    elif typeCurr == types[2]:\n",
    "        action_time = sessions[sessions['count'] == 0]\n",
    "    \n",
    "    action_time.groupby(['action_type']\n",
    "                     )['secs_elapsed'].sum().reset_index().plot('action_type', 'secs_elapsed',kind = 'bar',rot = 90)\n",
    "interact(actionTypeTime, typeCurr = widgets.Dropdown(options = {'All', 'Book', 'NDF'}))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## combine everything"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style>\n",
       "    .dataframe thead tr:only-child th {\n",
       "        text-align: right;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: left;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>0</th>\n",
       "      <th>1</th>\n",
       "      <th>2</th>\n",
       "      <th>3</th>\n",
       "      <th>4</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>user_id</th>\n",
       "      <td>00023iyk9l</td>\n",
       "      <td>001wyh0pz8</td>\n",
       "      <td>0028jgx1x1</td>\n",
       "      <td>002qnbzfs5</td>\n",
       "      <td>0035hobuyj</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-unknown-_x</th>\n",
       "      <td>0</td>\n",
       "      <td>6</td>\n",
       "      <td>1</td>\n",
       "      <td>184</td>\n",
       "      <td>9</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_request</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>booking_response</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>click</th>\n",
       "      <td>4</td>\n",
       "      <td>66</td>\n",
       "      <td>9</td>\n",
       "      <td>140</td>\n",
       "      <td>206</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>data</th>\n",
       "      <td>9</td>\n",
       "      <td>2</td>\n",
       "      <td>5</td>\n",
       "      <td>140</td>\n",
       "      <td>41</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>message_post</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>16</td>\n",
       "      <td>3</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>partner_callback</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>submit</th>\n",
       "      <td>0</td>\n",
       "      <td>3</td>\n",
       "      <td>1</td>\n",
       "      <td>15</td>\n",
       "      <td>4</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>view</th>\n",
       "      <td>21</td>\n",
       "      <td>8</td>\n",
       "      <td>15</td>\n",
       "      <td>216</td>\n",
       "      <td>55</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_destination_x</th>\n",
       "      <td>US</td>\n",
       "      <td>NDF</td>\n",
       "      <td>NDF</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>count_x</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>-unknown-_y</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>30</td>\n",
       "      <td>14</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Android App Unknown Phone/Tablet</th>\n",
       "      <td>0</td>\n",
       "      <td>90</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Android Phone</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Blackberry</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Chromebook</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Linux Desktop</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Mac Desktop</th>\n",
       "      <td>36</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>489</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Opera Phone</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Tablet</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Windows Desktop</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Windows Phone</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>iPad Tablet</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>iPhone</th>\n",
       "      <td>4</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>775</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>iPodtouch</th>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_destination_y</th>\n",
       "      <td>US</td>\n",
       "      <td>NDF</td>\n",
       "      <td>NDF</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>count_y</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>secs_elapsed</th>\n",
       "      <td>867896</td>\n",
       "      <td>282965</td>\n",
       "      <td>297010</td>\n",
       "      <td>6.48708e+06</td>\n",
       "      <td>5.72467e+06</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>country_destination</th>\n",
       "      <td>US</td>\n",
       "      <td>NDF</td>\n",
       "      <td>NDF</td>\n",
       "      <td>US</td>\n",
       "      <td>US</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>count</th>\n",
       "      <td>1</td>\n",
       "      <td>0</td>\n",
       "      <td>0</td>\n",
       "      <td>1</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>timeclass</th>\n",
       "      <td>(608254.0, 1277171.0]</td>\n",
       "      <td>(191518.0, 608254.0]</td>\n",
       "      <td>(191518.0, 608254.0]</td>\n",
       "      <td>(2474073.0, 38221363.0]</td>\n",
       "      <td>(2474073.0, 38221363.0]</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                      0                     1  \\\n",
       "user_id                                      00023iyk9l            001wyh0pz8   \n",
       "-unknown-_x                                           0                     6   \n",
       "booking_request                                       1                     0   \n",
       "booking_response                                      0                     0   \n",
       "click                                                 4                    66   \n",
       "data                                                  9                     2   \n",
       "message_post                                          1                     0   \n",
       "partner_callback                                      1                     0   \n",
       "submit                                                0                     3   \n",
       "view                                                 21                     8   \n",
       "country_destination_x                                US                   NDF   \n",
       "count_x                                               1                     0   \n",
       "-unknown-_y                                           0                     0   \n",
       "Android App Unknown Phone/Tablet                      0                    90   \n",
       "Android Phone                                         0                     0   \n",
       "Blackberry                                            0                     0   \n",
       "Chromebook                                            0                     0   \n",
       "Linux Desktop                                         0                     0   \n",
       "Mac Desktop                                          36                     0   \n",
       "Opera Phone                                           0                     0   \n",
       "Tablet                                                0                     0   \n",
       "Windows Desktop                                       0                     0   \n",
       "Windows Phone                                         0                     0   \n",
       "iPad Tablet                                           0                     0   \n",
       "iPhone                                                4                     0   \n",
       "iPodtouch                                             0                     0   \n",
       "country_destination_y                                US                   NDF   \n",
       "count_y                                               1                     0   \n",
       "secs_elapsed                                     867896                282965   \n",
       "country_destination                                  US                   NDF   \n",
       "count                                                 1                     0   \n",
       "timeclass                         (608254.0, 1277171.0]  (191518.0, 608254.0]   \n",
       "\n",
       "                                                     2  \\\n",
       "user_id                                     0028jgx1x1   \n",
       "-unknown-_x                                          1   \n",
       "booking_request                                      0   \n",
       "booking_response                                     0   \n",
       "click                                                9   \n",
       "data                                                 5   \n",
       "message_post                                         0   \n",
       "partner_callback                                     0   \n",
       "submit                                               1   \n",
       "view                                                15   \n",
       "country_destination_x                              NDF   \n",
       "count_x                                              0   \n",
       "-unknown-_y                                         30   \n",
       "Android App Unknown Phone/Tablet                     0   \n",
       "Android Phone                                        1   \n",
       "Blackberry                                           0   \n",
       "Chromebook                                           0   \n",
       "Linux Desktop                                        0   \n",
       "Mac Desktop                                          0   \n",
       "Opera Phone                                          0   \n",
       "Tablet                                               0   \n",
       "Windows Desktop                                      0   \n",
       "Windows Phone                                        0   \n",
       "iPad Tablet                                          0   \n",
       "iPhone                                               0   \n",
       "iPodtouch                                            0   \n",
       "country_destination_y                              NDF   \n",
       "count_y                                              0   \n",
       "secs_elapsed                                    297010   \n",
       "country_destination                                NDF   \n",
       "count                                                0   \n",
       "timeclass                         (191518.0, 608254.0]   \n",
       "\n",
       "                                                        3  \\\n",
       "user_id                                        002qnbzfs5   \n",
       "-unknown-_x                                           184   \n",
       "booking_request                                         1   \n",
       "booking_response                                        0   \n",
       "click                                                 140   \n",
       "data                                                  140   \n",
       "message_post                                           16   \n",
       "partner_callback                                        0   \n",
       "submit                                                 15   \n",
       "view                                                  216   \n",
       "country_destination_x                                  US   \n",
       "count_x                                                 1   \n",
       "-unknown-_y                                            14   \n",
       "Android App Unknown Phone/Tablet                        0   \n",
       "Android Phone                                           0   \n",
       "Blackberry                                              0   \n",
       "Chromebook                                              0   \n",
       "Linux Desktop                                           0   \n",
       "Mac Desktop                                             0   \n",
       "Opera Phone                                             0   \n",
       "Tablet                                                  0   \n",
       "Windows Desktop                                         0   \n",
       "Windows Phone                                           0   \n",
       "iPad Tablet                                             0   \n",
       "iPhone                                                775   \n",
       "iPodtouch                                               0   \n",
       "country_destination_y                                  US   \n",
       "count_y                                                 1   \n",
       "secs_elapsed                                  6.48708e+06   \n",
       "country_destination                                    US   \n",
       "count                                                   1   \n",
       "timeclass                         (2474073.0, 38221363.0]   \n",
       "\n",
       "                                                        4  \n",
       "user_id                                        0035hobuyj  \n",
       "-unknown-_x                                             9  \n",
       "booking_request                                         0  \n",
       "booking_response                                        0  \n",
       "click                                                 206  \n",
       "data                                                   41  \n",
       "message_post                                            3  \n",
       "partner_callback                                        0  \n",
       "submit                                                  4  \n",
       "view                                                   55  \n",
       "country_destination_x                                  US  \n",
       "count_x                                                 1  \n",
       "-unknown-_y                                             0  \n",
       "Android App Unknown Phone/Tablet                        0  \n",
       "Android Phone                                           0  \n",
       "Blackberry                                              0  \n",
       "Chromebook                                              0  \n",
       "Linux Desktop                                           0  \n",
       "Mac Desktop                                           489  \n",
       "Opera Phone                                             0  \n",
       "Tablet                                                  0  \n",
       "Windows Desktop                                         0  \n",
       "Windows Phone                                           0  \n",
       "iPad Tablet                                             0  \n",
       "iPhone                                                  0  \n",
       "iPodtouch                                               0  \n",
       "country_destination_y                                  US  \n",
       "count_y                                                 1  \n",
       "secs_elapsed                                  5.72467e+06  \n",
       "country_destination                                    US  \n",
       "count                                                   1  \n",
       "timeclass                         (2474073.0, 38221363.0]  "
      ]
     },
     "execution_count": 39,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "sessions_data = pd.merge(action_type,device_type,on='user_id',how='inner')\n",
    "sessions_data = pd.merge(sessions_data,grpby,on='user_id',how='inner')\n",
    "sessions_data.head().T"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "NDF with less time spent"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a12885310>"
      ]
     },
     "execution_count": 40,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAfAAAAFGCAYAAACVP82OAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nAAAAAElFTkSuQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a1e670690>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sessions_data.groupby(['country_destination']  \\\n",
    "                     )['secs_elapsed'].mean().reset_index().plot('country_destination', \\\n",
    "                                        'secs_elapsed',kind = 'bar', rot = 90, legend = False, \\\n",
    "                                         title = \"Time Elapsed(secs) Distribution by Destionation\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "if partner_callback exist, the destination is highly likely to be an English spoken country as most of the users are from US, it is easier to communicate with call (US, CA, GB)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a1383a910>"
      ]
     },
     "execution_count": 41,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAb0AAAFGCAYAAAD+e0GeAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\nQmCC\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a2182e310>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sessions_data.groupby(['country_destination']\n",
    "                     )['partner_callback'].mean().reset_index().plot('country_destination', \\\n",
    "                    'partner_callback',kind = 'bar', title = 'Avearge Callback Distribution by Destination' ,rot = 90)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "NDF with less message post"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 42,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x1a12e41cd0>"
      ]
     },
     "execution_count": 42,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAesAAAFGCAYAAABKcbeRAAAABHNCSVQICAgIfAhkiAAAAAlwSFlz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\ncyDIaHFdWgAAAABJRU5ErkJggg==\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x1a1f8d0d50>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sessions_data.groupby(['country_destination']\n",
    "                     )['message_post'].mean().reset_index().plot('country_destination', \\\n",
    "                    'message_post',kind = 'bar',title = 'Average Message Post Distrbution by Destionation' ,rot = 90)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 43,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "e52844d4c27e4791861124c6a07b5501",
       "version_major": 2,
       "version_minor": 0
      },
      "text/plain": [
       "A Jupyter Widget"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "<function __main__.actionTypeCounterVisual>"
      ]
     },
     "execution_count": 43,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "from matplotlib.colors import Colormap\n",
    "bar_order_origin = ['NDF','US','other','FR','IT','GB','ES','CA','DE','NL','AU','PT']\n",
    "cat_vars = ['partner_callback', 'message_post', 'data', 'view','booking_request','booking_response','submit', 'click']\n",
    "\n",
    "def actionTypeCounterVisual(NDF, US, other, FR, IT, GB, ES, CA, DE, NL, AU, PT):\n",
    "    '''\n",
    "    This function shows the distribution of the destionation countries regarding to \n",
    "                                        the action frequency of each action type.\n",
    "    *param NDF, US, other, FR, IT, GB, ES, CA, DE, NL, AU, PT decides whether this country \n",
    "                                        will be taken into consideration\n",
    "    '''\n",
    "    selections = [NDF, US, other, FR, IT, GB, ES, CA, DE, NL, AU, PT]\n",
    "    fig, ax5 = plt.subplots(nrows=4, ncols=2, figsize=(16, 16))\n",
    "    raw_Data = sessions_data\n",
    "    bar_order = ['NDF','US','other','FR','IT','GB','ES','CA','DE','NL','AU','PT']\n",
    "    for i in range(len(selections)):\n",
    "        if selections[i] == False:\n",
    "            raw_Data = raw_Data[raw_Data.country_destination != bar_order[i]]\n",
    "            bar_order[i] = None\n",
    "    \n",
    "    bar_order = list(filter(None, bar_order))\n",
    "    for i,var in enumerate(cat_vars):\n",
    "        pltCatVar(raw_Data, bar_order, var, ax5,i)\n",
    "    plt.tight_layout()\n",
    "    #ax5[0].set_position([box.x0, box.y0 + box.height * 0.4, box.width, box.height * 0.6])\n",
    "    ax5[3][1].legend(loc='upper center', bbox_to_anchor=(-0.15, -0.7), ncol=6)\n",
    "\n",
    "def pltCatVar(input_Data, bar_order, var,axis,ax_num):\n",
    "    '''\n",
    "    This is the plot function of country distribution regarding to the frequency of the given keywords\n",
    "    *param input_Data: the raw data to process with\n",
    "    *param bar_order: the order of the destination countries\n",
    "    *param var: the keyword of action type that we are interested in\n",
    "    *param axis, ax_num: parameters related to subplot\n",
    "    '''\n",
    "    ctab = pd.crosstab([input_Data[var]], input_Data.country_destination).apply(lambda x: x/x.sum(), axis=1)\n",
    "    temp = [ctab.index[0]] * (ctab.index.size / 5)\n",
    "    temp.extend([ctab.index[ctab.index.size * 1 / 5]] * (ctab.index.size * 2 / 5 - ctab.index.size / 5))\n",
    "    temp.extend([ctab.index[ctab.index.size * 2 / 5]] * (ctab.index.size * 3 / 5 - ctab.index.size * 2 / 5))\n",
    "    temp.extend([ctab.index[ctab.index.size * 3 / 5]] * (ctab.index.size * 4 / 5 - ctab.index.size *3 / 5))\n",
    "    temp.extend([ctab.index[ctab.index.size * 4 / 5]] * (ctab.index.size - ctab.index.size * 4 / 5))\n",
    "    ctab.index = temp\n",
    "    ctab = ctab.groupby(ctab.index).mean()\n",
    "    ax = ctab[bar_order].plot(kind='bar', stacked=True, ax=axis.reshape(-1)[ax_num],legend=False, \n",
    "                         colormap='terrain', title = var)\n",
    "    axis = []\n",
    "    for i in range(4):\n",
    "        axis.append(str(ctab.index[ctab.index.size * i / 5]) + \" to \" + str(ctab.index[ctab.index.size * (i + 1) / 5]))\n",
    "    axis.append(\"greater \" + str(ctab.index[ctab.index.size * 4 / 5]))\n",
    "    ax.set_xticklabels(axis, rotation=45)\n",
    "    if var == 'booking_response':\n",
    "        ax.set_xticklabels([0, 1])\n",
    "        \n",
    "\n",
    "interact(actionTypeCounterVisual, NDF = True, US = True, other = True, FR = True, IT = True, \\\n",
    "         GB = True, ES = True, CA = True, DE = True, NL = True, AU = True, PT = True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "The more times of book_request, the more likely to be others\n",
    "\n",
    "If not NDF, large data counts may be FR ,FR and IT has more message_post"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 2",
   "language": "python",
   "name": "python2"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 2
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython2",
   "version": "2.7.14"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}