2_Data_cleaning_en.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# Emoji Sentiment Analysis with Tweets\n",
    "        \n",
    "## step2-Data cleaning\n",
    "### 2.1-First EDA\n",
    "### 2.2-Data cleaning\n",
    "        \n",
    "    1. remove url\n",
    "    2. remove user names\n",
    "    3. remove punctuations\n",
    "    4. remove stopwords\n",
    "    5. lower the words\n",
    "    6. lemmatization\n",
    "    7. keep only english characters and emojis\n",
    "\n",
    "### 2.3-Extract pure emojis and text   \n",
    "And, we generate 3 new coloumns which:\n",
    "- remove all emojis\n",
    "        \n",
    "----\n",
    "- include all emojis\n",
    "        \n",
    "----\n",
    "- only include concerned emojis\n",
    "\n",
    "![](https://encrypted-tbn0.gstatic.com/images?q=tbn%3AANd9GcTigQWzoYCNiDyrz1BN4WTf2X2k9OZ_yvW-FsmcIMsdS9fppNmh)\n"
   ]
  },
  {
   "attachments": {
    "image.png": {
     "image/png": "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"
    }
   },
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.1 Exploratory Data Analysis\n",
    "\n",
    "Our dataset consists of Tweets which contain at least one of the following emoji\n",
    "The baseline selection of common-used emoji refers to [Ian D. Wood, Sebastian Ruder, 2016](https://www.researchgate.net/publication/321057905_Emoji_as_Emotion_Tags_for_Tweets)\n",
    "        \n",
    "The tag of emoji is besed on human intuition accroding to the original paper, instead of meaning in context.\n",
    "![image.png](attachment:image.png)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# required libraries\n",
    "import pandas as pd\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "import numpy as np\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.util import ngrams\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from collections import defaultdict\n",
    "from collections import  Counter\n",
    "plt.style.use('ggplot')\n",
    "stop=set(stopwords.words('english'))\n",
    "import re\n",
    "from nltk.tokenize import word_tokenize\n",
    "import gensim\n",
    "import string"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 29 emoji contains in the cluster joy\n",
      "joy ['😀', '😂', '😃', '😄', '😆', '😇', '😉', '😊', '😋', '😌', '😍', '😎', '😏', '🌞', '☺', '😘', '😜', '😝', '😛', '😺', '😸', '😹', '😻', '😼', '❤', '💖', '💕', '😁', '♥']\n",
      "There are 9 emoji contains in the cluster anger\n",
      "anger ['😬', '😠', '😐', '😑', '😠', '😡', '😖', '😤', '😾']\n",
      "There are 1 emoji contains in the cluster disgust\n",
      "disgust ['💩']\n",
      "There are 7 emoji contains in the cluster fear\n",
      "fear ['😅', '😦', '😧', '😱', '😨', '😰', '🙀']\n",
      "There are 12 emoji contains in the cluster sad\n",
      "sad ['😔', '😕', '☹', '😫', '😩', '😢', '😥', '😪', '😓', '😭', '😿', '💔']\n",
      "There are 4 emoji contains in the cluster surprise\n",
      "surprise ['😳', '😯', '😵', '😲']\n"
     ]
    }
   ],
   "source": [
    "# emoji lists\n",
    "\n",
    "joy = ['\\U0001F600', '\\U0001F602', '\\U0001F603', '\\U0001F604',\n",
    "          '\\U0001F606', '\\U0001F607', '\\U0001F609', '\\U0001F60A',\n",
    "          '\\U0001F60B', '\\U0001F60C', '\\U0001F60D', '\\U0001F60E',\n",
    "          '\\U0001F60F', '\\U0001F31E', '\\U0000263A', '\\U0001F618',\n",
    "          '\\U0001F61C', '\\U0001F61D', '\\U0001F61B', '\\U0001F63A',\n",
    "          '\\U0001F638', '\\U0001F639', '\\U0001F63B', '\\U0001F63C',\n",
    "          '\\U00002764', '\\U0001F496', '\\U0001F495', '\\U0001F601',\n",
    "          '\\U00002665']#joy\n",
    "\n",
    "anger = ['\\U0001F62C', '\\U0001F620', '\\U0001F610',\n",
    "          '\\U0001F611', '\\U0001F620', '\\U0001F621', '\\U0001F616',\n",
    "          '\\U0001F624', '\\U0001F63E']#anger\n",
    "disgust= ['\\U0001F4A9']#disgust\n",
    "fear = ['\\U0001F605', '\\U0001F626', '\\U0001F627', '\\U0001F631',\n",
    "          '\\U0001F628', '\\U0001F630', '\\U0001F640']#fear\n",
    "sad = ['\\U0001F614', '\\U0001F615', '\\U00002639', '\\U0001F62B',\n",
    "          '\\U0001F629', '\\U0001F622', '\\U0001F625', '\\U0001F62A',\n",
    "          '\\U0001F613', '\\U0001F62D', '\\U0001F63F', '\\U0001F494']#sad\n",
    "surp = ['\\U0001F633', '\\U0001F62F', '\\U0001F635', '\\U0001F632']#surprise\n",
    "\n",
    "emojis = {'joy':joy, 'anger':anger, 'disgust':disgust, 'fear':fear, 'sad':sad, 'surprise':surp}\n",
    "for i in emojis:\n",
    "    print('There are {} emoji contains in the cluster {}'.format(len(emojis[i]), i))\n",
    "    print(i,emojis[i])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "We crawed 366857 Tweets which contain at least one concerned emoji from 2022-04-15 to 2022-05-28:\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>time</th>\n",
       "      <th>tweets</th>\n",
       "      <th>source</th>\n",
       "      <th>entities</th>\n",
       "      <th>retweet</th>\n",
       "      <th>favorite</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1517890741070143488</td>\n",
       "      <td>2022-04-23 15:38:48+00:00</td>\n",
       "      <td>😂😂😂😂😂😅 Everyone would be awake by the time I g...</td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1517890740705460231</td>\n",
       "      <td>2022-04-23 15:38:48+00:00</td>\n",
       "      <td>RT @TheBigDome: Tag yourselves, we’re @imszmc ...</td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>355.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>1517890738675191809</td>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>@SnipesNFT @0xWave @BoredApeYC never unfortuna...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>3</td>\n",
       "      <td>1517890737639149568</td>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>RT @ESPNFC: FOUR GOALS FOR GABRIEL JESUS 😱 htt...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>229.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>4</td>\n",
       "      <td>1517890736544530432</td>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>RT @buitengebieden_: Kitten found the best spo...</td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [{'text': 'caturday', 'indices': ...</td>\n",
       "      <td>1022.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366852</th>\n",
       "      <td>385095</td>\n",
       "      <td>1528884215344795654</td>\n",
       "      <td>2022-05-23 23:42:56+00:00</td>\n",
       "      <td>RT @HotMessConserv2: A 13 year old boy is unde...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>6340.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366853</th>\n",
       "      <td>385096</td>\n",
       "      <td>1528884192733372417</td>\n",
       "      <td>2022-05-23 23:42:51+00:00</td>\n",
       "      <td>expose yourself :\\n\\n1. I would rather slengka...</td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366854</th>\n",
       "      <td>385097</td>\n",
       "      <td>1528884176363040773</td>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>I just want the People Above My Pay Grade to h...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366855</th>\n",
       "      <td>385098</td>\n",
       "      <td>1528884175302184960</td>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>@tonytaslucas @JPFinlayNBCS @john_keim That’s ...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366856</th>\n",
       "      <td>385099</td>\n",
       "      <td>1528884169660583938</td>\n",
       "      <td>2022-05-23 23:42:45+00:00</td>\n",
       "      <td>RT @MulhollandL0ver: You left Afghanistan afte...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>366857 rows × 9 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Unnamed: 0                   id                       time  \\\n",
       "0                0  1517890741070143488  2022-04-23 15:38:48+00:00   \n",
       "1                1  1517890740705460231  2022-04-23 15:38:48+00:00   \n",
       "2                2  1517890738675191809  2022-04-23 15:38:47+00:00   \n",
       "3                3  1517890737639149568  2022-04-23 15:38:47+00:00   \n",
       "4                4  1517890736544530432  2022-04-23 15:38:47+00:00   \n",
       "...            ...                  ...                        ...   \n",
       "366852      385095  1528884215344795654  2022-05-23 23:42:56+00:00   \n",
       "366853      385096  1528884192733372417  2022-05-23 23:42:51+00:00   \n",
       "366854      385097  1528884176363040773  2022-05-23 23:42:47+00:00   \n",
       "366855      385098  1528884175302184960  2022-05-23 23:42:47+00:00   \n",
       "366856      385099  1528884169660583938  2022-05-23 23:42:45+00:00   \n",
       "\n",
       "                                                   tweets  \\\n",
       "0       😂😂😂😂😂😅 Everyone would be awake by the time I g...   \n",
       "1       RT @TheBigDome: Tag yourselves, we’re @imszmc ...   \n",
       "2       @SnipesNFT @0xWave @BoredApeYC never unfortuna...   \n",
       "3       RT @ESPNFC: FOUR GOALS FOR GABRIEL JESUS 😱 htt...   \n",
       "4       RT @buitengebieden_: Kitten found the best spo...   \n",
       "...                                                   ...   \n",
       "366852  RT @HotMessConserv2: A 13 year old boy is unde...   \n",
       "366853  expose yourself :\\n\\n1. I would rather slengka...   \n",
       "366854  I just want the People Above My Pay Grade to h...   \n",
       "366855  @tonytaslucas @JPFinlayNBCS @john_keim That’s ...   \n",
       "366856  RT @MulhollandL0ver: You left Afghanistan afte...   \n",
       "\n",
       "                     source  \\\n",
       "0       Twitter for Android   \n",
       "1       Twitter for Android   \n",
       "2        Twitter for iPhone   \n",
       "3        Twitter for iPhone   \n",
       "4       Twitter for Android   \n",
       "...                     ...   \n",
       "366852   Twitter for iPhone   \n",
       "366853  Twitter for Android   \n",
       "366854   Twitter for iPhone   \n",
       "366855   Twitter for iPhone   \n",
       "366856   Twitter for iPhone   \n",
       "\n",
       "                                                 entities  retweet  favorite  \\\n",
       "0       {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "1       {'hashtags': [], 'symbols': [], 'user_mentions...    355.0       0.0   \n",
       "2       {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "3       {'hashtags': [], 'symbols': [], 'user_mentions...    229.0       0.0   \n",
       "4       {'hashtags': [{'text': 'caturday', 'indices': ...   1022.0       0.0   \n",
       "...                                                   ...      ...       ...   \n",
       "366852  {'hashtags': [], 'symbols': [], 'user_mentions...   6340.0       0.0   \n",
       "366853  {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "366854  {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       1.0   \n",
       "366855  {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "366856  {'hashtags': [], 'symbols': [], 'user_mentions...     25.0       0.0   \n",
       "\n",
       "           type  \n",
       "0          fear  \n",
       "1          fear  \n",
       "2          fear  \n",
       "3          fear  \n",
       "4          fear  \n",
       "...         ...  \n",
       "366852  disgust  \n",
       "366853  disgust  \n",
       "366854  disgust  \n",
       "366855  disgust  \n",
       "366856  disgust  \n",
       "\n",
       "[366857 rows x 9 columns]"
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "#load the crawed data\n",
    "tweet = pd.read_csv('tweets3.csv')\n",
    "tweet = tweet[tweet.type != 'type']#due to mistake when scraping, some rows are written as header repeatedly by mistake\n",
    "tweet = tweet.drop_duplicates()#drop duplicated rows\n",
    "print('We crawed {} Tweets which contain at least one concerned emoji from 2022-04-15 to 2022-05-28:'.format(len(tweet)))\n",
    "tweet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "iVBORw0KGgoAAAANSUhEUgAAAYsAAAEMCAYAAAA1VZrrAAAAOXRFWHRTb2Z0d2FyZQBNYXRwbG90bGliIHZlcnNpb24zLjMuMiwgaHR0cHM6Ly9tYXRwbG90bGliLm9yZy8vihELAAAACXBIWXMAAAsTAAALEwEAmpwYAAAixElEQVR4nO3dfViUdb7H8ffADLMaGGKMqHi8TptpK6WbuLtsNrjW8hDQ6hS7Jumuno6bu5ntJoXIwWy1dEOh7KC1W13lcSvyARQHsHJB006pl+cYK571bGKlLg+LiaAgM8z5o8s5ItrNkw7m5/UPze/+3eP32wzzuX9zz9yYPB6PBxERka/h5+sCRESk91NYiIiIIYWFiIgYUliIiIghhYWIiBhSWIiIiKEOhUVDQwOJiYl88cUXAOzatYukpCRiYmLIzs72zquoqMDhcBAbG8uCBQtwuVwAHDt2jJSUFOLi4pg9ezaNjY0A1NfXM2vWLOLj40lJSaGmpgaAs2fPkpqaSnx8PJMnT+Zvf/tbjzYtIiKdYxgW//3f/80DDzxAZWUlAE1NTaSnp5Obm4vT6aS8vJyysjIAUlNTyczMpKSkBI/HQ15eHgCLFi1i6tSpFBcXExERQW5uLgA5OTlERkZSVFREcnIyS5YsAWDNmjX06dOHoqIi0tPTmT9//uXoXUREOshsNCEvL4+FCxfyxBNPALB//36GDRvG0KFDAUhKSqK4uJibbrqJpqYmxowZA4DD4eCFF14gOTmZ3bt38+///u/e8QcffJDU1FRKS0tZu3YtAImJiTz99NO0tLRQWlrK3LlzARg3bhx1dXUcO3aMwYMHGzbU2tpKY2MjFosFk8nU+f8jIiLXII/HQ0tLC9dddx1+fu3XEYZhce5o/5zq6mpCQ0O9t202G1VVVe3GQ0NDqaqq4sSJEwQGBmI2m9uMX3hfZrOZwMBA6urqLnpff//73zsUFo2Njfz1r381nCciIu3dfPPNBAUFtRvv9Anu1tbWNkfsHo8Hk8l0yfFzP893qSN+j8eDn59fu33OjXeExWLpTDsiInKeS72GGq4sLhQWFuY9EQ1QU1ODzWZrN15bW4vNZiMkJIRTp07hdrvx9/f3zoevViW1tbWEhYXhcrlobGwkODiYgQMHUl1dzT/90z+1ua+OOBcyERERWK3WzrYnInJNam5upry8/JIH851eWYwePZrDhw9z5MgR3G43hYWF2O12hgwZgtVqZe/evQAUFBRgt9uxWCxERkbidDoByM/Px263AxAdHU1+fj4ATqeTyMhILBYL0dHRFBQUALBnzx6sVmuH3oISEZHLw9TRq85OnDiRN954g/DwcD788EOeffZZmpubiY6OZv78+ZhMJg4ePEhGRgYNDQ2MGjWKZ599loCAAI4ePUpaWhr/+Mc/GDRoECtWrOD666/nyy+/JC0tjc8//5ygoCCysrIIDw+nubmZzMxMysvLCQgIYPHixYwaNapDDZ1LR60sREQ6zui1s8NhcbVQWIiIdJ7Ra6e+wS0iIoYUFiIiYkhhISIihhQWIiJi6JoKC5e71dcldNrVWLOIfPN0+kt5VzOzvx+5W/f7uoxO+VXMbb4uQUTk2lpZiIhI1ygsRETEkMJCREQMKSxERMSQwkJERAwpLERExJDCQkREDCksRETEkMJCREQMKSxERMSQwkJERAwpLERExJDCQkREDCksRETEkMJCREQMKSxERMSQwkJERAwpLERExJDCQkREDCksRETEkMJCREQMKSxERMSQwkJERAwpLERExJDC4hukxd3q6xI65WqrV+RaZvZ1AdJzLP5+PLn2A1+X0WHLUsb7ugQR6SCtLERExJDCQkREDHUrLAoKCkhISCAhIYFly5YBsGvXLpKSkoiJiSE7O9s7t6KiAofDQWxsLAsWLMDlcgFw7NgxUlJSiIuLY/bs2TQ2NgJQX1/PrFmziI+PJyUlhZqamu6UKiIi3dDlsDhz5gxLlixhzZo1FBQUsGfPHrZt20Z6ejq5ubk4nU7Ky8spKysDIDU1lczMTEpKSvB4POTl5QGwaNEipk6dSnFxMREREeTm5gKQk5NDZGQkRUVFJCcns2TJkh5oV0REuqLLYeF2u2ltbeXMmTO4XC5cLheBgYEMGzaMoUOHYjabSUpKori4mKNHj9LU1MSYMWMAcDgcFBcX09LSwu7du4mNjW0zDlBaWkpSUhIAiYmJbN++nZaWlm62KyIiXdHlT0MFBgYyd+5c4uPj6dOnD+PGjaO6uprQ0FDvHJvNRlVVVbvx0NBQqqqqOHHiBIGBgZjN5jbjQJt9zGYzgYGB1NXVMXDgwK6WLCIiXdTlsDh48CDr16/nz3/+M0FBQcybN4/KykpMJpN3jsfjwWQy0draetHxcz/Pd+Ht8/fx8+v4Qqi8vLzd2NixYzu8f2+yd+/eDs27GvvraG8i4ltdDosPPviAqKgoBgwYAHz1FtIrr7yCv7+/d05NTQ02m42wsLA2J6hra2ux2WyEhIRw6tQp3G43/v7+3vnw1aqktraWsLAwXC4XjY2NBAcHd7i+iIgIrFZrV9vrVa7GEOiob3JvIleT5ubmix5kn9PlcxYjR45k165dnD59Go/Hw7Zt2xg9ejSHDx/myJEjuN1uCgsLsdvtDBkyBKvV6j2KLCgowG63Y7FYiIyMxOl0ApCfn4/dbgcgOjqa/Px8AJxOJ5GRkVgslq6WKyIi3dDllcX48eM5cOAADocDi8XCrbfeypw5c7jjjjuYM2cOzc3NREdHExcXB0BWVhYZGRk0NDQwatQopk+fDsDChQtJS0tj1apVDBo0iBUrVgAwd+5c0tLSSEhIICgoiKysrB5oV0REusLk8Xg8vi6iJ51bSl3qbajcrft9UFXX/Srmtk7N1+U+RKQrjF479Q1uERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMdStsNi2bRsOh4P4+HgWL14MwK5du0hKSiImJobs7Gzv3IqKChwOB7GxsSxYsACXywXAsWPHSElJIS4ujtmzZ9PY2AhAfX09s2bNIj4+npSUFGpqarpTqoiIdEOXw+Lzzz9n4cKF5ObmsmnTJg4cOEBZWRnp6enk5ubidDopLy+nrKwMgNTUVDIzMykpKcHj8ZCXlwfAokWLmDp1KsXFxURERJCbmwtATk4OkZGRFBUVkZyczJIlS3qgXRER6Youh8W7777LPffcQ1hYGBaLhezsbPr06cOwYcMYOnQoZrOZpKQkiouLOXr0KE1NTYwZMwYAh8NBcXExLS0t7N69m9jY2DbjAKWlpSQlJQGQmJjI9u3baWlp6Wa7IiLSFeau7njkyBEsFgsPP/wwx48fZ8KECQwfPpzQ0FDvHJvNRlVVFdXV1W3GQ0NDqaqq4sSJEwQGBmI2m9uMA232MZvNBAYGUldXx8CBA7tasoiIdFGXw8LtdrNnzx7WrFlD3759mT17Nt/61rcwmUzeOR6PB5PJRGtr60XHz/0834W3z9/Hz6/jC6Hy8vJ2Y2PHju3w/r3J3r17OzTvauyvo72JiG91OSxuuOEGoqKiCAkJAeDuu++muLgYf39/75yamhpsNhthYWFtTlDX1tZis9kICQnh1KlTuN1u/P39vfPhq1VJbW0tYWFhuFwuGhsbCQ4O7nB9ERERWK3WrrbXq1yNIdBR3+TeRK4mzc3NFz3IPqfL5yx+9KMf8cEHH1BfX4/b7WbHjh3ExcVx+PBhjhw5gtvtprCwELvdzpAhQ7Bard6jyIKCAux2OxaLhcjISJxOJwD5+fnY7XYAoqOjyc/PB8DpdBIZGYnFYulquSIi0g1dXlmMHj2ahx56iKlTp9LS0sIdd9zBAw88wI033sicOXNobm4mOjqauLg4ALKyssjIyKChoYFRo0Yxffp0ABYuXEhaWhqrVq1i0KBBrFixAoC5c+eSlpZGQkICQUFBZGVl9UC7IiLSFSaPx+PxdRE96dxS6lJvQ+Vu3e+DqrruVzG3dWr+k2s/uEyV9LxlKeM7Nf+sy02A2d94Yi9xtdUr1zaj184uryxErrQAsz8P5BT5uowOe/OxeF+XINJjdLkPERExpLAQERFDCgsRETGksBAREUMKC5Fe4GyLy9cldNrVWLN0nT4NJdILBFjMxMx/1ddldMrWZ2f6ugS5grSyEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQkcuuuaXF1yV02tVY8+Vk9nUBIvLNZ7VYmDBzga/L6JTSV5f4uoReRSsLERExpLAQERFDCgsRETGksBAREUMKCxERMdQjYbFs2TLS0tIA2LVrF0lJScTExJCdne2dU1FRgcPhIDY2lgULFuByuQA4duwYKSkpxMXFMXv2bBobGwGor69n1qxZxMfHk5KSQk1NTU+UKiIiXdDtsPjwww/ZuHEjAE1NTaSnp5Obm4vT6aS8vJyysjIAUlNTyczMpKSkBI/HQ15eHgCLFi1i6tSpFBcXExERQW5uLgA5OTlERkZSVFREcnIyS5boY2wiIr7SrbD48ssvyc7O5uGHHwZg//79DBs2jKFDh2I2m0lKSqK4uJijR4/S1NTEmDFjAHA4HBQXF9PS0sLu3buJjY1tMw5QWlpKUlISAImJiWzfvp0WfUlGRMQnuhUWmZmZ/OY3v6Ffv34AVFdXExoa6t1us9moqqpqNx4aGkpVVRUnTpwgMDAQs9ncZvzC+zKbzQQGBlJXV9edckVEpIu6/A3ud955h0GDBhEVFcWGDRsAaG1txWQyeed4PB5MJtMlx8/9PN+Ft8/fx8+v49lWXl7ebmzs2LEd3r832bt3b4fmXY39dbQ3+Gb3dzX2BurvWtLlsHA6ndTU1PCTn/yEkydPcvr0aY4ePYq/v793Tk1NDTabjbCwsDYnqGtra7HZbISEhHDq1Cncbjf+/v7e+fDVqqS2tpawsDBcLheNjY0EBwd3uL6IiAisVmtX2+tVrtZftI74JvcG6u9q903v73zNzc0XPcg+p8tvQ7322msUFhZSUFDAo48+ysSJE/njH//I4cOHOXLkCG63m8LCQux2O0OGDMFqtXpTuqCgALvdjsViITIyEqfTCUB+fj52ux2A6Oho8vPzga+CKTIyEovF0tVyRUSkG3r0QoJWq5WlS5cyZ84cmpubiY6OJi4uDoCsrCwyMjJoaGhg1KhRTJ8+HYCFCxeSlpbGqlWrGDRoECtWrABg7ty5pKWlkZCQQFBQEFlZWT1ZqoiIdEKPhIXD4cDhcAAQFRXFpk2b2s0ZOXIk69atazc+ZMgQ1qxZ0248ODiY1atX90R5IiLSTfoGt4iIGFJYiIiIIYWFiIgYUliIiIghhYWIiBhSWIiIiCGFhYiIGFJYiIiIIYWFiIgYUliIiIghhYWIiBhSWIiIiCGFhYiIGFJYiIiIIYWFiIgYUliIiIghhYWIiBhSWIiIiCGFhYiIGFJYiIh0U/PZs74uodM6W7P5MtUhInLNsAYE8MOEn/m6jE7ZteXtTs3XykJERAwpLERExJDCQkREDCksRETEkMJCREQMKSxERMSQwkJERAwpLERExJDCQkREDCksRETEkMJCREQMKSxERMSQwkJERAx1KyxefPFFEhISSEhI4Pe//z0Au3btIikpiZiYGLKzs71zKyoqcDgcxMbGsmDBAlwuFwDHjh0jJSWFuLg4Zs+eTWNjIwD19fXMmjWL+Ph4UlJSqKmp6U6pIiLSDV0Oi127dvHBBx+wceNG8vPz+ctf/kJhYSHp6enk5ubidDopLy+nrKwMgNTUVDIzMykpKcHj8ZCXlwfAokWLmDp1KsXFxURERJCbmwtATk4OkZGRFBUVkZyczJIlS3qgXRER6Youh0VoaChpaWkEBARgsVj49re/TWVlJcOGDWPo0KGYzWaSkpIoLi7m6NGjNDU1MWbMGAAcDgfFxcW0tLSwe/duYmNj24wDlJaWkpSUBEBiYiLbt2+npaWlm+2KiEhXdDkshg8f7n3xr6yspKioCJPJRGhoqHeOzWajqqqK6urqNuOhoaFUVVVx4sQJAgMDMZvNbcaBNvuYzWYCAwOpq6vrarkiItIN3f5LeYcOHeKXv/wlTzzxBP7+/lRWVnq3eTweTCYTra2tmEymduPnfp7vwtvn7+Pn1/FsKy8vbzc2duzYDu/fm+zdu7dD867G/jraG3yz+7saewP1d843vT/oZljs3buXRx99lPT0dBISEvj444/bnIiuqanBZrMRFhbWZry2thabzUZISAinTp3C7Xbj7+/vnQ9frUpqa2sJCwvD5XLR2NhIcHBwh2uLiIjAarV2p71e42p9InbEN7k3UH9Xu2upv+bm5oseZJ/T5behjh8/zq9//WuysrJISEgAYPTo0Rw+fJgjR47gdrspLCzEbrczZMgQrFarN8UKCgqw2+1YLBYiIyNxOp0A5OfnY7fbAYiOjiY/Px8Ap9NJZGQkFoulq+WKiEg3dHll8corr9Dc3MzSpUu9Y1OmTGHp0qXMmTOH5uZmoqOjiYuLAyArK4uMjAwaGhoYNWoU06dPB2DhwoWkpaWxatUqBg0axIoVKwCYO3cuaWlpJCQkEBQURFZWVnf6FBGRbuhyWGRkZJCRkXHRbZs2bWo3NnLkSNatW9dufMiQIaxZs6bdeHBwMKtXr+5qeSIi0oP0DW4RETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQwoLERExpLAQERFDCgsRETGksBAREUMKCxERMaSwEBERQ706LDZv3sw999xDTEwMa9eu9XU5IiLXLLOvC7iUqqoqsrOz2bBhAwEBAUyZMoXvf//73HTTTb4uTUTkmtNrVxa7du3iBz/4AcHBwfTt25fY2FiKi4t9XZaIyDWp164sqqurCQ0N9d622Wzs37/fcD+PxwPA2bNnL7o9wM/TMwVeIc3NzZ2a39diukyV9LzO9gbQz9prj2/a6Wx/wX0tl6mSy6Oz/fUP6nuZKrk8OttfSHC/y1TJ5XFhf+deM8+9hl7I5LnUFh9btWoVzc3NPPbYYwDk5eVRXl7O008//bX7nTp1ir/+9a9XoEIRkW+em2++maCgoHbjvXZlERYWxp49e7y3a2pqsNlshvtdd9113HzzzVgsFkymq+coW0TElzweDy0tLVx33XUX3d5rw+KHP/whK1eupK6ujj59+rB161Z+97vfGe7n5+d30VQUEZGv961vfeuS23ptWAwcOJDf/OY3TJ8+nZaWFu6//35uu+02X5clInJN6rXnLEREpPe4ej5aIiIiPqOwEBERQwoLERExpLAQERFDCgsRETGksBCvTz75hAULFvi6jF7hiy++YOLEib4uw9D8+fO56667KCws9HUpvUZVVRX/+q//6usy2klLS2P16tVXpLY///nPvPbaaz16n732exZy5d16663ceuutvi5DOmHjxo3s37+fgIAAX5fSawwcOJA//OEPvi7jomw22xWprby8vMfvU2FxAZfLxVNPPcWhQ4eora1lxIgRPP744zz++OMMHz6ciooKBgwYwPPPP09wcDBOp5MXXniBvn37csstt+B2u1m6dCn79+/n2Wefpampif79+7No0SKGDh3KtGnTuP766zl06BA5OTnccsstvm7Z66OPPuLFF1/k6aefJjMzky+//JK+ffuyYMECbrzxRu666y7ef/99AgMD+eKLL5g1axZOp9PXZX+tv//978ybN4/Tp0/j5+dHRkYGx48f57XXXqOpqYmzZ8/yzDPPcPvtt3PgwAHvymrkyJE+rtzYww8/jMfjITk5mRkzZvD666/T2trKqFGjWLhwIVarlf/4j/+goKCAM2fOYLFYWL58OTfeeCMTJ07ktttuo6Kigj/96U8MGDDgitV9scfkt7/9LW+88Qbh4eHe5+GaNWva/b7MnDmTH//4x+zbt4/rrruOrKwswsPD2/Tz3HPP8dhjj7Ft2zY2b97MH//4R/z9/QkPD+e5557DarXy8ssvU1RUhNvtZvz48aSmpvb45YE8Hg9Lly6ltLQUm82G2+3me9/7HhMnTvza2pYvX05JSQn9+/cnNDSUiRMn8r3vfY/p06ezbds2AFauXAl89RxIT0/n0KFDAEydOpXbb7+dt956C4DBgwdz33339Ug/ehvqAvv27cNisfD222/z7rvvcurUKcrKyjh48CAzZsygsLCQfv36sXnzZurq6njmmWd4/fXXWbduHSdPngS+unpjRkYGy5cvZ+PGjcyYMYN/+7d/8/4bI0aMoKSkpFcFxflSU1OZNm0amzdvZv78+cydO5eAgAAmTJjgvUx8fn4+kyZN8m2hHbBu3TomTJjAhg0bePTRR9m9ezdvvfUWq1evZtOmTTz00EO8/PLLADz55JPMmzePjRs3Eh4e7uPKja1evRqArKws8vLyeOuttygoKGDAgAG88sorNDQ08N5777FmzRoKCwuZMGFCmz8iZrfbKSkpuaJBAe0fk717937t/PN/X+rq6vjud7/L5s2bSUhIYPHixd555/oJCQnxjuXk5PDqq6+yYcMGhgwZwqeffsr27dspLy9n3bp15OfnU1VVxaZNm3q8z5KSEg4cOEBhYSHPP/88n332WZvtF6tt27Zt7N27l8LCQl5++WUOHDjwtf/Gvn37OHnyJPn5+bz00kvs2bOHm266iSlTpjBlypQeCwrQyqKdcePGERwczNq1a/n000+prKzk9OnTDBgwgO985zsADB8+nJMnT7Jnzx6++93vMnDgQAAmTZrEe++9R2VlJZ9//jmzZ8/23m9DQ4P3v3vzZUsaGxv54osviImJAWDMmDFcf/31fPrpp9x3332sXLmS+++/n8LCQl5//XUfV2ssKiqKOXPmUFFRQXR0NNOnT+eBBx5g27ZtHD58mI8//hg/Pz/q6uqorq7mjjvuAMDhcLB+/XofV98xH330EUeOHOGnP/0pAC0tLXznO98hMDCQ5cuXs2XLFiorK9mxY0ebA5TRo0f7pN4LH5MHH3zwa/8S5vm/L1ar1XuQMnnyZFasWOHddrF+fvSjH/HAAw9w9913Exsbyy233MKmTZvYv38/DocDgKamJgYPHtxD3f2/jz/+mJiYGCwWCyEhIdjtdsPa1q9fT3x8PAEBAQQEBHD33Xd/7b8xfPhwDh8+zL/8y79gt9t54okneryPcxQWF3j//fd54YUXmD59Og6HgxMnTjB48GCsVqt3jslkwuPx4OfnR2tra7v7aG1tJTw8nIKCAgDcbje1tbXe7V93sS5fu9jVXzweD263m3HjxlFdXc3WrVsJDw/3hmRvNnbsWLZs2UJpaSlOp5N33nmHmpoa7r33XsaNG8eIESNYu3at9zE9x9/f34dVd47b7SY+Pp6MjAzgq8B3u90cP36cadOm8eCDD2K327nhhhuoqKjw7nf+c/pKuvAx2bhxI/D/zz2Xy9Vm/vm/L35+ft63i1pbW9s8ThfrJyMjg4MHD1JWVkZqaiqPPPIIbrebn//858yYMQOA+vr6y/J4X/icMpvbvtxerLZLvaZceF8ulwuz2Uz//v3ZsmULO3fupKysjMmTJ7Nly5Ye7wX0NlQ7H374IfHx8dx3333069ePjz76CLfbfdG5t99+O5988gnV1dV4PB6cTicmk4kbb7zRu/IAWL9+PfPmzbuSbXRZYGAg4eHhbN26FYD/+q//ora2luHDh2MymZg0aRKLFy/2HpX1dr///e/ZtGkTkydPJjMzk48//hiTycTDDz/M97//fd59913cbjf9+/dn8ODBlJaWAlxVny4618c//vEPPB4PTz31FK+//jqffPIJw4YN4xe/+AW33nor77333iWfy1fShY/JgQMH6N+/P//7v/8LfHXAdilnzpzxvm+/YcOGdkfr53O5XMTExNC/f39++ctf8pOf/ISKigp+8IMfUFBQQGNjIy6Xi1//+teUlJT0bJN8tYIqKiri7NmznDx5kh07dhjW9sMf/pCtW7dy9uxZGhoaKC0txWQy0a9fP7788kvq6uo4e/as977ef/99UlNTmTBhAhkZGfTt25fjx4/j7+/fLnS7SyuLCyQnJzNv3jy2bNmCxWLh9ttv56OPPrro3JCQEDIyMpg5cyYBAQGEh4fTr18/AgICeP7551myZAnNzc0EBgaybNmyK9xJ1z333HM89dRTrFy5EovFwsqVK72ftklISODVV181XB73FtOmTePxxx9nw4YN+Pv789JLL7Fp0ybi4+MxmUyMHz/e+575c889x/z588nJyWHMmDG+LbwTRo4cySOPPMLPf/5zWltbueWWW5g1axYul4s333yTe+65B4/Hw7hx47wnQn3pwsdk2bJlmEwmfve73/Hiiy8yfvz4r92/uLiY7OxsbDbb1/5emc1mHn30UWbOnInVamXAgAEsXbqUAQMGcPDgQX7605/idru58847mTx5ck+3yd13380nn3xCYmIiN9xwA9/+9rc7VNu+ffuYPHky119/PTabDavVSlBQEA899BD3338/YWFh3k8t2u12tm7dSkJCAlarlXvvvZcRI0ZQX1/Pk08+yQ033MC0adN6pB9ddbYbTpw4wZo1a7zLx8WLFzNs2LAee3CutPfee4933nmHl1566aLbW1tbefPNNzl8+LD3LQ+RK2nEiBH8z//8j6/LuGz27dtHZWUlkydPpqWlhZ/97Gc888wzveLTeVpZdENwcDD19fUkJibi7+/PqFGjvCcZrzZOp5MlS5aQnp5+yTmPPPIIx48f55VXXrmClYlcO/75n/+ZF198kddeew2Px8OkSZN6RVCAVhYiItIBOsEtIiKGFBYiImJIYSEiIoYUFiI9ZObMmdTV1fm6DJHLQmEh0kN27tzp6xJELhuFhUgPmD9/PgB33nknI0aM8F6y4cyZM0RFRVFXV8fEiRNZvnw5DoeDH//4x/zpT3/y7r9t2zaSk5OZNGkSU6ZMYd++fT7pQ+RS9D0LkR7w7LPPsmHDBnbs2MGMGTPYsWMH0dHRbNmyhaioKO+VUE+ePMn69eupqqpi0qRJjB07FqvVSnZ2Nm+88Qb9+/fn0KFDzJgxg61bt9K3b18fdybyFYWFSA9LSUkhLy+P6Oho3n777TZXAp06dSomk4mwsDDuvPNOdu7cidVqpbq6ml/84hfeeSaTic8++6zXfCFLRGEh0sOSkpJYsWIF//mf/8np06cZN26cd9v5Vx5tbW31XmU0KiqKnJwc77bjx49js9muZNkiX0vnLER6yLkrffbp04d7772X9PR0pkyZ0mZOfn4+AMeOHWPnzp3Y7XaioqLYuXMnf/vb3wAoKyvj3nvvpamp6Uq3IHJJutyHSA/57W9/y1/+8hdWrlxJa2srU6ZMYceOHQQFBQF4//TnZ599RlNTE7/61a9ITEwEoKioiNWrV+PxeDCbzaSnpxMZGenLdkTaUFiI9DCPx8Mf/vAHjh49yqJFi7zjEydO5Pnnn/deXlrkaqJzFiI97K677sJms5Gbm+vrUkR6jFYWIiJiSCe4RUTEkMJCREQMKSxERMSQwkJERAwpLERExJDCQkREDP0fZNweXd15YMMAAAAASUVORK5CYII=\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "type\n",
       "anger       95850\n",
       "joy         84076\n",
       "sad         57790\n",
       "fear        55269\n",
       "surprise    45178\n",
       "disgust     28694\n",
       "dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# 2.1.1 data clusters contained in our dataset\n",
    "# we considered the number of emoji in each cluster, as well as the balance of samples\n",
    "sns.set(style=\"whitegrid\", color_codes=True)\n",
    "data = tweet.groupby(\"type\").size().sort_values(ascending = False)\n",
    "pal = sns.color_palette(\"Blues_d\", len(data))\n",
    "sns.barplot(x=data.index, y=data, palette=pal)\n",
    "\n",
    "plt.show()\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# 2.1.2 number of words in tweets\n",
    "lengths = [len(t.split(' ')) for t in tweet['tweets']]\n",
    "\n",
    "plt.figure(figsize=(10,4))\n",
    "plt.grid(axis = 'x')\n",
    "plt.hist(lengths, bins = len(set(lengths)))\n",
    "plt.show()\n",
    "\n",
    "# most tweets contain words from 0 to 40\n",
    "# Intuitively, this dosen't mean the longer tweets are outliers, we don't drop them so far\n",
    "# but when taining models in the next section, we should consider the length distribution"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 2.1.3 find the common stopwords and punctuations in tweets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "corpus=[]\n",
    "for x in tweet['tweets'].str.split():\n",
    "    for i in x:\n",
    "        corpus.append(i)\n",
    "\n",
    "dic=defaultdict(int)\n",
    "for word in corpus:\n",
    "    if word in stop:\n",
    "        dic[word]+=1\n",
    "\n",
    "top=sorted(dic.items(), key=lambda x:x[1],reverse=True)[:20]\n",
    "pal = sns.color_palette(\"Blues_d\", len(top))\n",
    "x,y=zip(*top)\n",
    "plt.figure(figsize=(15,4))\n",
    "plt.grid(axis = 'x')\n",
    "plt.bar(x,y,color=pal)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1080x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# punctuations\n",
    "dic=defaultdict(int)\n",
    "import string\n",
    "special = string.punctuation\n",
    "for i in (corpus):\n",
    "    if i in special:\n",
    "        dic[i]+=1\n",
    "        \n",
    "x,y=zip(*dic.items())\n",
    "pal = sns.color_palette(\"Blues_d\", len(dic.items()))\n",
    "plt.figure(figsize=(15,4))\n",
    "plt.grid(axis = 'x')\n",
    "plt.bar(x,y,color=pal)\n",
    "plt.show()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.2 Data Cleaning\n",
    "We include 6 data cleaning steps:\n",
    "        \n",
    "    1. remove url\n",
    "    2. remove user names\n",
    "    3. remove punctuations\n",
    "    4. remove stopwords\n",
    "    5. lower the words\n",
    "    6. lemmatization\n",
    "\n",
    "Then, we construct a new column which remove all emoji for the future modeling"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [],
   "source": [
    "################"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "# 2.2.1 data cleaning\n",
    "# a new dataframe, keep the original dataset\n",
    "df = pd.read_csv('tweets3.csv')\n",
    "df = df[df.type != 'type']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "RT @buitengebieden_: Kitten found the best spot.. 😅\n",
      "\n",
      "#caturday https://t.co/6SgrPCAv4A\n",
      "---------------\n",
      "Removing url...\n",
      "RT @buitengebieden_: Kitten found the best spot.. 😅\n",
      "\n",
      "#caturday \n"
     ]
    }
   ],
   "source": [
    "# i. remove url\n",
    "def remove_URL(text):\n",
    "    url = re.compile(r'https?://\\S+|www\\.\\S+')\n",
    "    return url.sub(r'',text)\n",
    "\n",
    "example = df['tweets'][4]\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Removing url...')\n",
    "example = remove_URL(example)\n",
    "print(example)\n",
    "\n",
    "df['tweets']=df['tweets'].apply(lambda x : remove_URL(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "RT @buitengebieden_: Kitten found the best spot.. 😅\n",
      "\n",
      "#caturday \n",
      "---------------\n",
      "Removing user names...\n",
      " Kitten found the best spot.. 😅\n",
      "\n",
      "#caturday \n"
     ]
    }
   ],
   "source": [
    "# ii. remove user names\n",
    "def remove_name(text):\n",
    "    name=re.compile(r'R?T?.*?@\\S+')\n",
    "    return name.sub(r'',text)\n",
    "\n",
    "example = df['tweets'][4]\n",
    "# example = \"RT @XtraSweetBobby: It's crazy we'll have a nice day and then like 3 days in a row where the wind is just blowing hard… \"\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Removing user names...')\n",
    "example = remove_name(example)\n",
    "print(example)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['tweets']=df['tweets'].apply(lambda x : remove_name(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      " Kitten found the best spot.. 😅\n",
      "\n",
      "#caturday \n",
      "---------------\n",
      "Removing punctuations...\n",
      " Kitten found the best spot 😅\n",
      "\n",
      "caturday \n"
     ]
    }
   ],
   "source": [
    "# iii. remove punctuations\n",
    "def remove_punct(text):\n",
    "    table=str.maketrans('','',string.punctuation)\n",
    "    return text.translate(table)\n",
    "\n",
    "example = df['tweets'][4]\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Removing punctuations...')\n",
    "example = remove_punct(example)\n",
    "print(example)\n",
    "\n",
    "df['tweets']=df['tweets'].apply(lambda x : remove_punct(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      " Kitten found the best spot 😅\n",
      "\n",
      "caturday \n",
      "---------------\n",
      "Removing stopwords...\n",
      "Kitten best spot 😅 caturday\n"
     ]
    }
   ],
   "source": [
    "# iv. remove stopwords\n",
    "from gensim.parsing.preprocessing import remove_stopwords\n",
    "example = df['tweets'][4]\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Removing stopwords...')\n",
    "example = remove_stopwords(example)\n",
    "print(example)\n",
    "\n",
    "df['tweets']=df['tweets'].apply(lambda x : remove_stopwords(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "Kitten best spot 😅 caturday\n",
      "---------------\n",
      "Lowering ...\n",
      "kitten best spot 😅 caturday\n"
     ]
    }
   ],
   "source": [
    "# v. lower\n",
    "example = df['tweets'][4]\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Lowering ...')\n",
    "example = example.lower()\n",
    "print(example)\n",
    "\n",
    "df['tweets']=df['tweets'].apply(lambda x : x.lower())"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "kitten best spot 😅 caturday tweets looks\n",
      "---------------\n",
      "Lemmatizating ...\n",
      "kitten best spot 😅 caturday tweet look\n"
     ]
    }
   ],
   "source": [
    "#v.i. lemmatization\n",
    "\n",
    "from textblob import TextBlob\n",
    "\n",
    "def lemm(text):\n",
    "    textTB = TextBlob(text)\n",
    "    words = textTB.words\n",
    "    words_lemmatized = words.lemmatize()\n",
    "    return ' '.join(words_lemmatized)\n",
    "\n",
    "example = df['tweets'][4]+' tweets'+' looks'\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Lemmatizating ...')\n",
    "example = lemm(example)\n",
    "print(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "CPU times: user 1min 7s, sys: 351 ms, total: 1min 7s\n",
      "Wall time: 1min 8s\n"
     ]
    }
   ],
   "source": [
    "%time df['tweets']=df['tweets'].apply(lambda x : lemm(x))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# v.ii. keep only english characters and emojis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "kitten best spot 😅 caturday我 们 لغة عربية 123 😅😅\n",
      "---------------\n",
      "Extracting English words and emojis ...\n",
      "kitten best spot  😅 \n"
     ]
    },
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "[nltk_data] Error loading words: <urlopen error [Errno 61] Connection\n",
      "[nltk_data]     refused>\n"
     ]
    }
   ],
   "source": [
    "import emoji\n",
    "import nltk \n",
    "nltk.download('words')\n",
    "words = set(nltk.corpus.words.words())\n",
    "\n",
    "def keepengemoji(text):\n",
    "    ls = []\n",
    "    for w in text.split(' '):\n",
    "        if w in words:\n",
    "            ls.append(w)\n",
    "        elif w in emoji.UNICODE_EMOJI['en']:\n",
    "            w = ' '+w+' '\n",
    "            ls.append(w)\n",
    "        else:\n",
    "            continue\n",
    "    return ' '.join(ls)\n",
    "\n",
    "example = df['tweets'][4]+'我 们 لغة عربية 123 😅😅'\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Extracting English words and emojis ...')\n",
    "example = keepengemoji(example)\n",
    "print(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "df['tweets']=df['tweets'].apply(lambda x : keepengemoji(x))"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### 2.3. Extract pure text and emojis"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "################\n",
    "# i. Form a new column (pure text)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "kitten best spot  😅 \n",
      "---------------\n",
      "Extracting pure text ...\n",
      "kitten best spot   \n"
     ]
    }
   ],
   "source": [
    "# function of removing emoji\n",
    "import emoji\n",
    "\n",
    "def remove_emojis(text):\n",
    "  return ''.join(c for c in text if c not in emoji.UNICODE_EMOJI['en'])\n",
    "\n",
    "example = df['tweets'][4]\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Extracting pure text ...')\n",
    "example = remove_emojis(example)\n",
    "print(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['everyone awake time i room',\n",
       " '',\n",
       " 'unfortunately',\n",
       " 'four goal for   ',\n",
       " 'kitten best spot   ',\n",
       " '',\n",
       " 'they looking wrong i hate',\n",
       " 'you guy play time',\n",
       " 'challenge community voting is officially you we second look',\n",
       " '']"
      ]
     },
     "execution_count": 24,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "puretext = [remove_emojis(t) for t in df['tweets']]\n",
    "puretext[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "kitten best spot  😅 😅😅😅😅😅\n",
      "---------------\n",
      "Extracting emoji ...\n",
      "😅 😅 😅 😅 😅 😅\n"
     ]
    }
   ],
   "source": [
    "# function of extracting emoji\n",
    "import emoji\n",
    "\n",
    "def extract_emojis(text):\n",
    "  return ' '.join(c for c in text if c in emoji.UNICODE_EMOJI['en'])\n",
    "\n",
    "example = df['tweets'][4]+'😅😅😅😅😅'\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Extracting emoji ...')\n",
    "example = extract_emojis(example)\n",
    "print(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['', '', '', '😱', '😅', '', '', '', '', '']"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "emojis = [extract_emojis(t) for t in df['tweets']]\n",
    "emojis[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "same play different jordan  😱 😱🥶 💨 🖼🥳\n",
      "---------------\n",
      "Extracting concerned emoji ...\n",
      "😱 😱\n"
     ]
    }
   ],
   "source": [
    "# function of get concerned emojis\n",
    "allemoji = joy + anger + disgust + fear + sad + surp\n",
    "\n",
    "def concern_emojis(text):\n",
    "  return ' '.join(c for c in text if c in allemoji)\n",
    "\n",
    "example = df['tweets'][59]+'😱🥶 💨 🖼🥳'\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Extracting concerned emoji ...')\n",
    "example = concern_emojis(example)\n",
    "print(example)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['', '', '', '😱', '😅', '', '', '', '', '']"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "c_emojis = [concern_emojis(t) for t in df['tweets']]\n",
    "c_emojis[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>time</th>\n",
       "      <th>tweets</th>\n",
       "      <th>puretext</th>\n",
       "      <th>emojis</th>\n",
       "      <th>c_emojis</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2022-04-23 15:38:48+00:00</td>\n",
       "      <td>everyone awake time i room</td>\n",
       "      <td>everyone awake time i room</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2022-04-23 15:38:48+00:00</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>unfortunately</td>\n",
       "      <td>unfortunately</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>four goal for  😱</td>\n",
       "      <td>four goal for</td>\n",
       "      <td>😱</td>\n",
       "      <td>😱</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>kitten best spot  😅</td>\n",
       "      <td>kitten best spot</td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366852</th>\n",
       "      <td>2022-05-23 23:42:56+00:00</td>\n",
       "      <td>a year old boy sexual harassment investigation...</td>\n",
       "      <td>a year old boy sexual harassment investigation...</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366853</th>\n",
       "      <td>2022-05-23 23:42:51+00:00</td>\n",
       "      <td>expose i maybe black p</td>\n",
       "      <td>expose i maybe black p</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366854</th>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>i want people above my pay grade energy giving...</td>\n",
       "      <td>i want people above my pay grade energy giving...</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366855</th>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>that s fan endure going it s  💩</td>\n",
       "      <td>that s fan endure going it s</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366856</th>\n",
       "      <td>2022-05-23 23:42:45+00:00</td>\n",
       "      <td>you left control you there met</td>\n",
       "      <td>you left control you there met</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>366857 rows × 6 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             time  \\\n",
       "0       2022-04-23 15:38:48+00:00   \n",
       "1       2022-04-23 15:38:48+00:00   \n",
       "2       2022-04-23 15:38:47+00:00   \n",
       "3       2022-04-23 15:38:47+00:00   \n",
       "4       2022-04-23 15:38:47+00:00   \n",
       "...                           ...   \n",
       "366852  2022-05-23 23:42:56+00:00   \n",
       "366853  2022-05-23 23:42:51+00:00   \n",
       "366854  2022-05-23 23:42:47+00:00   \n",
       "366855  2022-05-23 23:42:47+00:00   \n",
       "366856  2022-05-23 23:42:45+00:00   \n",
       "\n",
       "                                                   tweets  \\\n",
       "0                              everyone awake time i room   \n",
       "1                                                           \n",
       "2                                           unfortunately   \n",
       "3                                       four goal for  😱    \n",
       "4                                    kitten best spot  😅    \n",
       "...                                                   ...   \n",
       "366852  a year old boy sexual harassment investigation...   \n",
       "366853                             expose i maybe black p   \n",
       "366854  i want people above my pay grade energy giving...   \n",
       "366855                   that s fan endure going it s  💩    \n",
       "366856                     you left control you there met   \n",
       "\n",
       "                                                 puretext emojis c_emojis  \\\n",
       "0                              everyone awake time i room                   \n",
       "1                                                                           \n",
       "2                                           unfortunately                   \n",
       "3                                        four goal for         😱        😱   \n",
       "4                                     kitten best spot         😅        😅   \n",
       "...                                                   ...    ...      ...   \n",
       "366852  a year old boy sexual harassment investigation...                   \n",
       "366853                             expose i maybe black p                   \n",
       "366854  i want people above my pay grade energy giving...                   \n",
       "366855                    that s fan endure going it s         💩        💩   \n",
       "366856                     you left control you there met                   \n",
       "\n",
       "           type  \n",
       "0          fear  \n",
       "1          fear  \n",
       "2          fear  \n",
       "3          fear  \n",
       "4          fear  \n",
       "...         ...  \n",
       "366852  disgust  \n",
       "366853  disgust  \n",
       "366854  disgust  \n",
       "366855  disgust  \n",
       "366856  disgust  \n",
       "\n",
       "[366857 rows x 6 columns]"
      ]
     },
     "execution_count": 29,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# a new df\n",
    "df2 = df\n",
    "df2['puretext'] = puretext\n",
    "df2['emojis'] = emojis\n",
    "df2['c_emojis'] = c_emojis\n",
    "df2[['time','tweets','puretext','emojis','c_emojis','type']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Unnamed: 0</th>\n",
       "      <th>id</th>\n",
       "      <th>time</th>\n",
       "      <th>tweets</th>\n",
       "      <th>source</th>\n",
       "      <th>entities</th>\n",
       "      <th>retweet</th>\n",
       "      <th>favorite</th>\n",
       "      <th>type</th>\n",
       "      <th>puretext</th>\n",
       "      <th>emojis</th>\n",
       "      <th>c_emojis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>0</td>\n",
       "      <td>1517890741070143488</td>\n",
       "      <td>2022-04-23 15:38:48+00:00</td>\n",
       "      <td>everyone awake time i room</td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "      <td>everyone awake time i room</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1</td>\n",
       "      <td>1517890740705460231</td>\n",
       "      <td>2022-04-23 15:38:48+00:00</td>\n",
       "      <td></td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>355.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2</td>\n",
       "      <td>1517890738675191809</td>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>unfortunately</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "      <td>unfortunately</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>5</td>\n",
       "      <td>1517890736141733888</td>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td></td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>6</td>\n",
       "      <td>1517890735772774400</td>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>they looking wrong i hate</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>fear</td>\n",
       "      <td>they looking wrong i hate</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366850</th>\n",
       "      <td>385093</td>\n",
       "      <td>1528884298560204800</td>\n",
       "      <td>2022-05-23 23:43:16+00:00</td>\n",
       "      <td>your source statement or making</td>\n",
       "      <td>Twitter for iPad</td>\n",
       "      <td>{'hashtags': [{'text': 'WhitePowerBarbie', 'in...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>5.0</td>\n",
       "      <td>disgust</td>\n",
       "      <td>your source statement or making</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366852</th>\n",
       "      <td>385095</td>\n",
       "      <td>1528884215344795654</td>\n",
       "      <td>2022-05-23 23:42:56+00:00</td>\n",
       "      <td>a year old boy sexual harassment investigation...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>6340.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "      <td>a year old boy sexual harassment investigation...</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366853</th>\n",
       "      <td>385096</td>\n",
       "      <td>1528884192733372417</td>\n",
       "      <td>2022-05-23 23:42:51+00:00</td>\n",
       "      <td>expose i maybe black p</td>\n",
       "      <td>Twitter for Android</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "      <td>expose i maybe black p</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366854</th>\n",
       "      <td>385097</td>\n",
       "      <td>1528884176363040773</td>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>i want people above my pay grade energy giving...</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>0.0</td>\n",
       "      <td>1.0</td>\n",
       "      <td>disgust</td>\n",
       "      <td>i want people above my pay grade energy giving...</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366856</th>\n",
       "      <td>385099</td>\n",
       "      <td>1528884169660583938</td>\n",
       "      <td>2022-05-23 23:42:45+00:00</td>\n",
       "      <td>you left control you there met</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>{'hashtags': [], 'symbols': [], 'user_mentions...</td>\n",
       "      <td>25.0</td>\n",
       "      <td>0.0</td>\n",
       "      <td>disgust</td>\n",
       "      <td>you left control you there met</td>\n",
       "      <td></td>\n",
       "      <td></td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>206816 rows × 12 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "        Unnamed: 0                   id                       time  \\\n",
       "0                0  1517890741070143488  2022-04-23 15:38:48+00:00   \n",
       "1                1  1517890740705460231  2022-04-23 15:38:48+00:00   \n",
       "2                2  1517890738675191809  2022-04-23 15:38:47+00:00   \n",
       "5                5  1517890736141733888  2022-04-23 15:38:47+00:00   \n",
       "6                6  1517890735772774400  2022-04-23 15:38:47+00:00   \n",
       "...            ...                  ...                        ...   \n",
       "366850      385093  1528884298560204800  2022-05-23 23:43:16+00:00   \n",
       "366852      385095  1528884215344795654  2022-05-23 23:42:56+00:00   \n",
       "366853      385096  1528884192733372417  2022-05-23 23:42:51+00:00   \n",
       "366854      385097  1528884176363040773  2022-05-23 23:42:47+00:00   \n",
       "366856      385099  1528884169660583938  2022-05-23 23:42:45+00:00   \n",
       "\n",
       "                                                   tweets  \\\n",
       "0                              everyone awake time i room   \n",
       "1                                                           \n",
       "2                                           unfortunately   \n",
       "5                                                           \n",
       "6                               they looking wrong i hate   \n",
       "...                                                   ...   \n",
       "366850                    your source statement or making   \n",
       "366852  a year old boy sexual harassment investigation...   \n",
       "366853                             expose i maybe black p   \n",
       "366854  i want people above my pay grade energy giving...   \n",
       "366856                     you left control you there met   \n",
       "\n",
       "                     source  \\\n",
       "0       Twitter for Android   \n",
       "1       Twitter for Android   \n",
       "2        Twitter for iPhone   \n",
       "5        Twitter for iPhone   \n",
       "6        Twitter for iPhone   \n",
       "...                     ...   \n",
       "366850     Twitter for iPad   \n",
       "366852   Twitter for iPhone   \n",
       "366853  Twitter for Android   \n",
       "366854   Twitter for iPhone   \n",
       "366856   Twitter for iPhone   \n",
       "\n",
       "                                                 entities  retweet  favorite  \\\n",
       "0       {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "1       {'hashtags': [], 'symbols': [], 'user_mentions...    355.0       0.0   \n",
       "2       {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "5       {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "6       {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "...                                                   ...      ...       ...   \n",
       "366850  {'hashtags': [{'text': 'WhitePowerBarbie', 'in...      0.0       5.0   \n",
       "366852  {'hashtags': [], 'symbols': [], 'user_mentions...   6340.0       0.0   \n",
       "366853  {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       0.0   \n",
       "366854  {'hashtags': [], 'symbols': [], 'user_mentions...      0.0       1.0   \n",
       "366856  {'hashtags': [], 'symbols': [], 'user_mentions...     25.0       0.0   \n",
       "\n",
       "           type                                           puretext emojis  \\\n",
       "0          fear                         everyone awake time i room          \n",
       "1          fear                                                             \n",
       "2          fear                                      unfortunately          \n",
       "5          fear                                                             \n",
       "6          fear                          they looking wrong i hate          \n",
       "...         ...                                                ...    ...   \n",
       "366850  disgust                    your source statement or making          \n",
       "366852  disgust  a year old boy sexual harassment investigation...          \n",
       "366853  disgust                             expose i maybe black p          \n",
       "366854  disgust  i want people above my pay grade energy giving...          \n",
       "366856  disgust                     you left control you there met          \n",
       "\n",
       "       c_emojis  \n",
       "0                \n",
       "1                \n",
       "2                \n",
       "5                \n",
       "6                \n",
       "...         ...  \n",
       "366850           \n",
       "366852           \n",
       "366853           \n",
       "366854           \n",
       "366856           \n",
       "\n",
       "[206816 rows x 12 columns]"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# # some emojis in tweets are removed due to text processing\n",
    "df2.query('emojis == \"\"')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "@XtraSweetBobby It's crazy we'll have a nice day and then like 3 days in a row where the wind is just blowing hard… https://t.co/EdQaWQ67u7\n",
      "it crazy nice day like day row wind blowing\n",
      "it crazy nice day like day row wind blowing\n",
      "---------\n",
      "we now delete these samples\n"
     ]
    }
   ],
   "source": [
    "# example of mistook data\n",
    "i = 17\n",
    "print(tweet.loc[i]['tweets'])\n",
    "print(df.loc[i]['tweets'])\n",
    "print(df2.loc[i]['tweets'])\n",
    "\n",
    "print('---------')\n",
    "print('we now delete these samples')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>time</th>\n",
       "      <th>tweets</th>\n",
       "      <th>puretext</th>\n",
       "      <th>emojis</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>four goal for  😱</td>\n",
       "      <td>four goal for</td>\n",
       "      <td>😱</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>kitten best spot  😅</td>\n",
       "      <td>kitten best spot</td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2022-04-23 15:38:46+00:00</td>\n",
       "      <td>🙀  when cleaning window remember</td>\n",
       "      <td>when cleaning window remember</td>\n",
       "      <td>🙀</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2022-04-23 15:38:45+00:00</td>\n",
       "      <td>diamond dragon field day making fun  😅</td>\n",
       "      <td>diamond dragon field day making fun</td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2022-04-23 15:38:45+00:00</td>\n",
       "      <td>😅</td>\n",
       "      <td></td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366839</th>\n",
       "      <td>2022-05-23 23:43:58+00:00</td>\n",
       "      <td>t  💩  talk</td>\n",
       "      <td>t    talk</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366845</th>\n",
       "      <td>2022-05-23 23:43:45+00:00</td>\n",
       "      <td>you expect b  💩   😖  look  😡  black face</td>\n",
       "      <td>you expect b       look    black face</td>\n",
       "      <td>💩 😖 😡</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366846</th>\n",
       "      <td>2022-05-23 23:43:42+00:00</td>\n",
       "      <td>💩  never use crap</td>\n",
       "      <td>never use crap</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366851</th>\n",
       "      <td>2022-05-23 23:43:06+00:00</td>\n",
       "      <td>what  💩</td>\n",
       "      <td>what</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366855</th>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>that s fan endure going it s  💩</td>\n",
       "      <td>that s fan endure going it s</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>160041 rows × 5 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             time                                    tweets  \\\n",
       "3       2022-04-23 15:38:47+00:00                         four goal for  😱    \n",
       "4       2022-04-23 15:38:47+00:00                      kitten best spot  😅    \n",
       "11      2022-04-23 15:38:46+00:00          🙀  when cleaning window remember   \n",
       "13      2022-04-23 15:38:45+00:00   diamond dragon field day making fun  😅    \n",
       "14      2022-04-23 15:38:45+00:00                                        😅    \n",
       "...                           ...                                       ...   \n",
       "366839  2022-05-23 23:43:58+00:00                                t  💩  talk   \n",
       "366845  2022-05-23 23:43:45+00:00  you expect b  💩   😖  look  😡  black face   \n",
       "366846  2022-05-23 23:43:42+00:00                         💩  never use crap   \n",
       "366851  2022-05-23 23:43:06+00:00                                  what  💩    \n",
       "366855  2022-05-23 23:42:47+00:00          that s fan endure going it s  💩    \n",
       "\n",
       "                                      puretext emojis     type  \n",
       "3                             four goal for         😱     fear  \n",
       "4                          kitten best spot         😅     fear  \n",
       "11               when cleaning window remember      🙀     fear  \n",
       "13      diamond dragon field day making fun         😅     fear  \n",
       "14                                                  😅     fear  \n",
       "...                                        ...    ...      ...  \n",
       "366839                               t    talk      💩  disgust  \n",
       "366845   you expect b       look    black face  💩 😖 😡  disgust  \n",
       "366846                          never use crap      💩  disgust  \n",
       "366851                                 what         💩  disgust  \n",
       "366855         that s fan endure going it s         💩  disgust  \n",
       "\n",
       "[160041 rows x 5 columns]"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# we need to drop these invalid samples\n",
    "df3 = df2[df2.emojis != '']\n",
    "df3[['time','tweets','puretext','emojis','type']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>tweets</th>\n",
       "      <th>puretext</th>\n",
       "      <th>emojis</th>\n",
       "      <th>c_emojis</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>four goal for  😱</td>\n",
       "      <td>four goal for</td>\n",
       "      <td>😱</td>\n",
       "      <td>😱</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>kitten best spot  😅</td>\n",
       "      <td>kitten best spot</td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>🙀  when cleaning window remember</td>\n",
       "      <td>when cleaning window remember</td>\n",
       "      <td>🙀</td>\n",
       "      <td>🙀</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>diamond dragon field day making fun  😅</td>\n",
       "      <td>diamond dragon field day making fun</td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>😅</td>\n",
       "      <td></td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366839</th>\n",
       "      <td>t  💩  talk</td>\n",
       "      <td>t    talk</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366845</th>\n",
       "      <td>you expect b  💩   😖  look  😡  black face</td>\n",
       "      <td>you expect b       look    black face</td>\n",
       "      <td>💩 😖 😡</td>\n",
       "      <td>💩 😖 😡</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366846</th>\n",
       "      <td>💩  never use crap</td>\n",
       "      <td>never use crap</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366851</th>\n",
       "      <td>what  💩</td>\n",
       "      <td>what</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366855</th>\n",
       "      <td>that s fan endure going it s  💩</td>\n",
       "      <td>that s fan endure going it s</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>160041 rows × 4 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                          tweets  \\\n",
       "3                              four goal for  😱    \n",
       "4                           kitten best spot  😅    \n",
       "11              🙀  when cleaning window remember   \n",
       "13       diamond dragon field day making fun  😅    \n",
       "14                                            😅    \n",
       "...                                          ...   \n",
       "366839                                t  💩  talk   \n",
       "366845  you expect b  💩   😖  look  😡  black face   \n",
       "366846                         💩  never use crap   \n",
       "366851                                  what  💩    \n",
       "366855          that s fan endure going it s  💩    \n",
       "\n",
       "                                      puretext emojis c_emojis  \n",
       "3                             four goal for         😱        😱  \n",
       "4                          kitten best spot         😅        😅  \n",
       "11               when cleaning window remember      🙀        🙀  \n",
       "13      diamond dragon field day making fun         😅        😅  \n",
       "14                                                  😅        😅  \n",
       "...                                        ...    ...      ...  \n",
       "366839                               t    talk      💩        💩  \n",
       "366845   you expect b       look    black face  💩 😖 😡    💩 😖 😡  \n",
       "366846                          never use crap      💩        💩  \n",
       "366851                                 what         💩        💩  \n",
       "366855         that s fan endure going it s         💩        💩  \n",
       "\n",
       "[160041 rows x 4 columns]"
      ]
     },
     "execution_count": 33,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3[['tweets','puretext','emojis','c_emojis']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Example of original Tweet:\n",
      "four goal for  😱 \n",
      "---------------\n",
      "Spliting emojis...\n",
      "four goal for   😱  \n"
     ]
    }
   ],
   "source": [
    "###\n",
    "# split multi-emojis in tweets\n",
    "import emoji\n",
    "def emosplit(text):\n",
    "    words = []\n",
    "    for c in text:\n",
    "        if c in emoji.UNICODE_EMOJI['en']:\n",
    "            c = ' '+c+' '\n",
    "            words.append(c)\n",
    "        else:\n",
    "            words.append(c)\n",
    "    return ''.join(words)\n",
    "\n",
    "example = df2['tweets'][3]\n",
    "print('Example of original Tweet:')\n",
    "print(example)\n",
    "print('---------------')\n",
    "print('Spliting emojis...')\n",
    "example = emosplit(example)\n",
    "print(example)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['four goal for   😱  ',\n",
       " 'kitten best spot   😅  ',\n",
       " '  🙀   when cleaning window remember',\n",
       " 'diamond dragon field day making fun   😅  ',\n",
       " '  😅  ',\n",
       " 'quick giveaway sorry late   😅   winner peso like tweet following',\n",
       " 'same play different jordan   😱  ',\n",
       " '  😮   big news   😮   the team officially begun build phase we pushing',\n",
       " '  💙   purposely singing badly   ❤ ️  somebody bring guy   😅  ',\n",
       " 'statue   😅  ']"
      ]
     },
     "execution_count": 35,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "stext = [emosplit(t) for t in df3['tweets']]\n",
    "stext[:10]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "<ipython-input-36-05627c022991>:1: SettingWithCopyWarning: \n",
      "A value is trying to be set on a copy of a slice from a DataFrame.\n",
      "Try using .loc[row_indexer,col_indexer] = value instead\n",
      "\n",
      "See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy\n",
      "  df3['stweets'] = stext\n"
     ]
    },
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>time</th>\n",
       "      <th>stweets</th>\n",
       "      <th>tweets</th>\n",
       "      <th>puretext</th>\n",
       "      <th>emojis</th>\n",
       "      <th>c_emojis</th>\n",
       "      <th>type</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>four goal for   😱</td>\n",
       "      <td>four goal for  😱</td>\n",
       "      <td>four goal for</td>\n",
       "      <td>😱</td>\n",
       "      <td>😱</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>2022-04-23 15:38:47+00:00</td>\n",
       "      <td>kitten best spot   😅</td>\n",
       "      <td>kitten best spot  😅</td>\n",
       "      <td>kitten best spot</td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>11</th>\n",
       "      <td>2022-04-23 15:38:46+00:00</td>\n",
       "      <td>🙀   when cleaning window remember</td>\n",
       "      <td>🙀  when cleaning window remember</td>\n",
       "      <td>when cleaning window remember</td>\n",
       "      <td>🙀</td>\n",
       "      <td>🙀</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>13</th>\n",
       "      <td>2022-04-23 15:38:45+00:00</td>\n",
       "      <td>diamond dragon field day making fun   😅</td>\n",
       "      <td>diamond dragon field day making fun  😅</td>\n",
       "      <td>diamond dragon field day making fun</td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>14</th>\n",
       "      <td>2022-04-23 15:38:45+00:00</td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "      <td></td>\n",
       "      <td>😅</td>\n",
       "      <td>😅</td>\n",
       "      <td>fear</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366839</th>\n",
       "      <td>2022-05-23 23:43:58+00:00</td>\n",
       "      <td>t   💩   talk</td>\n",
       "      <td>t  💩  talk</td>\n",
       "      <td>t    talk</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366845</th>\n",
       "      <td>2022-05-23 23:43:45+00:00</td>\n",
       "      <td>you expect b   💩     😖   look   😡   black face</td>\n",
       "      <td>you expect b  💩   😖  look  😡  black face</td>\n",
       "      <td>you expect b       look    black face</td>\n",
       "      <td>💩 😖 😡</td>\n",
       "      <td>💩 😖 😡</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366846</th>\n",
       "      <td>2022-05-23 23:43:42+00:00</td>\n",
       "      <td>💩   never use crap</td>\n",
       "      <td>💩  never use crap</td>\n",
       "      <td>never use crap</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366851</th>\n",
       "      <td>2022-05-23 23:43:06+00:00</td>\n",
       "      <td>what   💩</td>\n",
       "      <td>what  💩</td>\n",
       "      <td>what</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>366855</th>\n",
       "      <td>2022-05-23 23:42:47+00:00</td>\n",
       "      <td>that s fan endure going it s   💩</td>\n",
       "      <td>that s fan endure going it s  💩</td>\n",
       "      <td>that s fan endure going it s</td>\n",
       "      <td>💩</td>\n",
       "      <td>💩</td>\n",
       "      <td>disgust</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>160041 rows × 7 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                             time  \\\n",
       "3       2022-04-23 15:38:47+00:00   \n",
       "4       2022-04-23 15:38:47+00:00   \n",
       "11      2022-04-23 15:38:46+00:00   \n",
       "13      2022-04-23 15:38:45+00:00   \n",
       "14      2022-04-23 15:38:45+00:00   \n",
       "...                           ...   \n",
       "366839  2022-05-23 23:43:58+00:00   \n",
       "366845  2022-05-23 23:43:45+00:00   \n",
       "366846  2022-05-23 23:43:42+00:00   \n",
       "366851  2022-05-23 23:43:06+00:00   \n",
       "366855  2022-05-23 23:42:47+00:00   \n",
       "\n",
       "                                               stweets  \\\n",
       "3                                  four goal for   😱     \n",
       "4                               kitten best spot   😅     \n",
       "11                   🙀   when cleaning window remember   \n",
       "13           diamond dragon field day making fun   😅     \n",
       "14                                                 😅     \n",
       "...                                                ...   \n",
       "366839                                    t   💩   talk   \n",
       "366845  you expect b   💩     😖   look   😡   black face   \n",
       "366846                              💩   never use crap   \n",
       "366851                                      what   💩     \n",
       "366855              that s fan endure going it s   💩     \n",
       "\n",
       "                                          tweets  \\\n",
       "3                              four goal for  😱    \n",
       "4                           kitten best spot  😅    \n",
       "11              🙀  when cleaning window remember   \n",
       "13       diamond dragon field day making fun  😅    \n",
       "14                                            😅    \n",
       "...                                          ...   \n",
       "366839                                t  💩  talk   \n",
       "366845  you expect b  💩   😖  look  😡  black face   \n",
       "366846                         💩  never use crap   \n",
       "366851                                  what  💩    \n",
       "366855          that s fan endure going it s  💩    \n",
       "\n",
       "                                      puretext emojis c_emojis     type  \n",
       "3                             four goal for         😱        😱     fear  \n",
       "4                          kitten best spot         😅        😅     fear  \n",
       "11               when cleaning window remember      🙀        🙀     fear  \n",
       "13      diamond dragon field day making fun         😅        😅     fear  \n",
       "14                                                  😅        😅     fear  \n",
       "...                                        ...    ...      ...      ...  \n",
       "366839                               t    talk      💩        💩  disgust  \n",
       "366845   you expect b       look    black face  💩 😖 😡    💩 😖 😡  disgust  \n",
       "366846                          never use crap      💩        💩  disgust  \n",
       "366851                                 what         💩        💩  disgust  \n",
       "366855         that s fan endure going it s         💩        💩  disgust  \n",
       "\n",
       "[160041 rows x 7 columns]"
      ]
     },
     "execution_count": 36,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df3['stweets'] = stext\n",
    "df3[['time','stweets','tweets','puretext','emojis','c_emojis','type']]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "data": {
      "text/plain": [
       "type\n",
       "anger       39476\n",
       "joy         34393\n",
       "fear        27920\n",
       "sad         23144\n",
       "surprise    22607\n",
       "disgust     12501\n",
       "dtype: int64"
      ]
     },
     "execution_count": 37,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "################\n",
    "# 2.2.3 Look at the parsed dataset\n",
    "\n",
    "sns.set(style=\"whitegrid\", color_codes=True)\n",
    "data = df3.groupby(\"type\").size().sort_values(ascending = False)\n",
    "pal = sns.color_palette(\"Blues_d\", len(data))\n",
    "sns.barplot(x=data.index, y=data, palette=pal)\n",
    "\n",
    "plt.show()\n",
    "data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 38,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# number of words in tweets\n",
    "lengths = [len(t.split(' ')) for t in df3['tweets']]\n",
    "\n",
    "plt.figure(figsize=(10,4))\n",
    "plt.grid(axis = 'x')\n",
    "plt.hist(lengths, bins = len(set(lengths)))\n",
    "plt.show()\n",
    "\n",
    "# after data cleaning,\n",
    "# most tweets contain words from 0 to 25"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 39,
   "metadata": {},
   "outputs": [],
   "source": [
    "# save new csvs\n",
    "df3.to_csv('tweets_cleaned_drop.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "# upload to github version\n",
    "df4 = df3[['time','tweets','stweets','puretext','emojis','c_emojis','type']]\n",
    "df4.to_csv('data_tweets_cleaned_en.csv')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.8.5"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}