Donald Trump Twitter Sentiment Analysis using TextBlob.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tweepy"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "Consumer_Key = ''\n",
    "Consumer_Secret = ''\n",
    "\n",
    "Access_Token = ''\n",
    "Access_Secret = ''"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "def twitter_setup():\n",
    "    \"\"\"\n",
    "    Utility function to setup the Twitter's API\n",
    "    with our access keys provided.\n",
    "    \"\"\"\n",
    "    # Authentication and access using keys:\n",
    "    auth = tweepy.OAuthHandler(Consumer_Key, Consumer_Secret)\n",
    "    auth.set_access_token(Access_Token, Access_Secret)\n",
    "\n",
    "    # Return API with authentication:\n",
    "    api = tweepy.API(auth)\n",
    "    return api"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of tweets extracted: 200.\n",
      "\n",
      "5 recent tweets: \n",
      "\n",
      "....Also, why is Mueller only appointing Angry Dems, some of whom have worked for Crooked Hillary, others, includin… https://t.co/hX52XFQ91d\n",
      "\n",
      "Is Robert Mueller ever going to release his conflicts of interest with respect to President Trump, including the fa… https://t.co/t0GHh6hsEg\n",
      "\n",
      "There is No Collusion! The Robert Mueller Rigged Witch Hunt, headed now by 17 (increased from 13, including an Obam… https://t.co/mtUqvuvzZD\n",
      "\n",
      "...and the Amazon Washington Post do nothing but write bad stories even on very positive achievements - and they will never change!\n",
      "\n",
      "...dying newspaper industry. No matter how much they try to distract and cover it up, our country is making great p… https://t.co/UVMw8JEf8i\n",
      "\n"
     ]
    }
   ],
   "source": [
    "extractor = twitter_setup()\n",
    "\n",
    "tweets = extractor.user_timeline(screen_name=\"realDonaldTrump\", count=300)\n",
    "\n",
    "print (\"Number of tweets extracted: {}.\\n\".format(len(tweets)))\n",
    "print (\"5 recent tweets: \\n\")\n",
    "for tweet in tweets[:5]:\n",
    "    print (tweet.text)\n",
    "    print()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import numpy as np\n",
    "import matplotlib as plt\n",
    "import seaborn as sns\n",
    "\n",
    "%matplotlib inline"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Tweet</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Congressman David Kustoff has been a champion ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Private business investment has surged from 1....</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>The @USNavy’s first female Admiral, Alene Duer...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>We have accomplished an economic turnaround of...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>I am thrilled to announce that in the second q...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               Tweet\n",
       "0  Congressman David Kustoff has been a champion ...\n",
       "1  Private business investment has surged from 1....\n",
       "2  The @USNavy’s first female Admiral, Alene Duer...\n",
       "3  We have accomplished an economic turnaround of...\n",
       "4  I am thrilled to announce that in the second q..."
      ]
     },
     "execution_count": 6,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data = pd.DataFrame(data=[tweet.text for tweet in tweets], columns=['Tweet'])\n",
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['len']= np.array([len(tweet.text) for tweet in tweets])\n",
    "data['ID']= np.array([tweet.id  for tweet in tweets])\n",
    "data['Date']= np.array([tweet.created_at  for tweet in tweets])\n",
    "data['Scource']= np.array([tweet.source  for tweet in tweets])\n",
    "data['Likes']= np.array([tweet.favorite_count  for tweet in tweets])\n",
    "data['RTs']= np.array([tweet.retweet_count  for tweet in tweets])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Tweet</th>\n",
       "      <th>len</th>\n",
       "      <th>ID</th>\n",
       "      <th>Date</th>\n",
       "      <th>Scource</th>\n",
       "      <th>Likes</th>\n",
       "      <th>RTs</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Congressman David Kustoff has been a champion ...</td>\n",
       "      <td>140</td>\n",
       "      <td>1022936690267176971</td>\n",
       "      <td>2018-07-27 20:08:06</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>17773</td>\n",
       "      <td>4618</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Private business investment has surged from 1....</td>\n",
       "      <td>140</td>\n",
       "      <td>1022925366669856769</td>\n",
       "      <td>2018-07-27 19:23:07</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>22009</td>\n",
       "      <td>5969</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>The @USNavy’s first female Admiral, Alene Duer...</td>\n",
       "      <td>140</td>\n",
       "      <td>1022908672756801536</td>\n",
       "      <td>2018-07-27 18:16:46</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>24173</td>\n",
       "      <td>5744</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>We have accomplished an economic turnaround of...</td>\n",
       "      <td>92</td>\n",
       "      <td>1022879441767288834</td>\n",
       "      <td>2018-07-27 16:20:37</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>35861</td>\n",
       "      <td>9474</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>I am thrilled to announce that in the second q...</td>\n",
       "      <td>140</td>\n",
       "      <td>1022877332858650624</td>\n",
       "      <td>2018-07-27 16:12:14</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>49944</td>\n",
       "      <td>12849</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               Tweet  len  \\\n",
       "0  Congressman David Kustoff has been a champion ...  140   \n",
       "1  Private business investment has surged from 1....  140   \n",
       "2  The @USNavy’s first female Admiral, Alene Duer...  140   \n",
       "3  We have accomplished an economic turnaround of...   92   \n",
       "4  I am thrilled to announce that in the second q...  140   \n",
       "\n",
       "                    ID                Date             Scource  Likes    RTs  \n",
       "0  1022936690267176971 2018-07-27 20:08:06  Twitter for iPhone  17773   4618  \n",
       "1  1022925366669856769 2018-07-27 19:23:07  Twitter for iPhone  22009   5969  \n",
       "2  1022908672756801536 2018-07-27 18:16:46  Twitter for iPhone  24173   5744  \n",
       "3  1022879441767288834 2018-07-27 16:20:37  Twitter for iPhone  35861   9474  \n",
       "4  1022877332858650624 2018-07-27 16:12:14  Twitter for iPhone  49944  12849  "
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "330847"
      ]
     },
     "execution_count": 25,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fav = np.max(data['Likes'])\n",
    "fav"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "67"
      ]
     },
     "execution_count": 26,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "fav_tweet"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Most liked tweet: \n",
      " To Iranian President Rouhani: NEVER, EVER THREATEN THE UNITED STATES AGAIN OR YOU WILL SUFFER CONSEQUENCES THE LIKE… https://t.co/xUdD66vfBW\n",
      "Most RT tweet: \n",
      " RT @realDonaldTrump: They just didn't get it, but they do now! https://t.co/9T50NupkDy\n"
     ]
    }
   ],
   "source": [
    "fav = np.max(data['Likes'])\n",
    "rt= np.max(data['RTs'])\n",
    "\n",
    "fav_tweet= data[data.Likes==fav].index[0]\n",
    "rt_tweet= data[data.RTs==rt].index[0]\n",
    "\n",
    "print (('Most liked tweet: \\n') , (data['Tweet'][fav_tweet]))\n",
    "print (('Most RT tweet: \\n'), (data['Tweet'][rt_tweet]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "likes= pd.Series(data=data['Likes'].values, index=data['Date'])\n",
    "RTs= pd.Series(data=data['RTs'].values, index=data['Date'])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "<matplotlib.axes._subplots.AxesSubplot at 0x226414cc828>"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<matplotlib.figure.Figure at 0x2263dcfb5c0>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "likes.plot()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "from textblob import TextBlob\n",
    "import re"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [],
   "source": [
    "def clean_data(tweet):\n",
    "    return ' '.join(re.sub(\"(@[A-Za-z0-9]+)|([^0-9A-Za-z \\t])|(\\w+:\\/\\/\\S+)\", \" \", tweet).split())\n",
    "def analyze_text(tweet):\n",
    "    analysis = TextBlob(clean_data(tweet))\n",
    "    if analysis.sentiment.polarity > 0:\n",
    "        return 1\n",
    "    elif analysis.sentiment.polarity == 0:\n",
    "        return 0\n",
    "    else: return -1\n",
    "    "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Congressman David Kustoff has been a champion for the Trump Agenda - I greatly appreciate his support. David is str… https://t.co/b69nex9NfW'"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data['Tweet'][0]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "0.0"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "a=clean_data(data['Tweet'][1])\n",
    "b=TextBlob(a)\n",
    "b.sentiment.polarity"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "array(<generator object <genexpr> at 0x00000226429D5CA8>, dtype=object)"
      ]
     },
     "execution_count": 17,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "b=np.array(analyze_text(tweet) for tweet in data['Tweet'] )\n",
    "b"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'Congressman David Kustoff has been a champion for the Trump Agenda I greatly appreciate his support David is str'"
      ]
     },
     "execution_count": 18,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "clean_data(data['Tweet'][0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "data['SA']= np.array([analyze_text(tweet) for tweet in data['Tweet']])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>Tweet</th>\n",
       "      <th>len</th>\n",
       "      <th>ID</th>\n",
       "      <th>Date</th>\n",
       "      <th>Scource</th>\n",
       "      <th>Likes</th>\n",
       "      <th>RTs</th>\n",
       "      <th>SA</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>Congressman David Kustoff has been a champion ...</td>\n",
       "      <td>140</td>\n",
       "      <td>1022936690267176971</td>\n",
       "      <td>2018-07-27 20:08:06</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>17773</td>\n",
       "      <td>4618</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Private business investment has surged from 1....</td>\n",
       "      <td>140</td>\n",
       "      <td>1022925366669856769</td>\n",
       "      <td>2018-07-27 19:23:07</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>22009</td>\n",
       "      <td>5969</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>The @USNavy’s first female Admiral, Alene Duer...</td>\n",
       "      <td>140</td>\n",
       "      <td>1022908672756801536</td>\n",
       "      <td>2018-07-27 18:16:46</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>24173</td>\n",
       "      <td>5744</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>We have accomplished an economic turnaround of...</td>\n",
       "      <td>92</td>\n",
       "      <td>1022879441767288834</td>\n",
       "      <td>2018-07-27 16:20:37</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>35861</td>\n",
       "      <td>9474</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>I am thrilled to announce that in the second q...</td>\n",
       "      <td>140</td>\n",
       "      <td>1022877332858650624</td>\n",
       "      <td>2018-07-27 16:12:14</td>\n",
       "      <td>Twitter for iPhone</td>\n",
       "      <td>49944</td>\n",
       "      <td>12849</td>\n",
       "      <td>1</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                               Tweet  len  \\\n",
       "0  Congressman David Kustoff has been a champion ...  140   \n",
       "1  Private business investment has surged from 1....  140   \n",
       "2  The @USNavy’s first female Admiral, Alene Duer...  140   \n",
       "3  We have accomplished an economic turnaround of...   92   \n",
       "4  I am thrilled to announce that in the second q...  140   \n",
       "\n",
       "                    ID                Date             Scource  Likes    RTs  \\\n",
       "0  1022936690267176971 2018-07-27 20:08:06  Twitter for iPhone  17773   4618   \n",
       "1  1022925366669856769 2018-07-27 19:23:07  Twitter for iPhone  22009   5969   \n",
       "2  1022908672756801536 2018-07-27 18:16:46  Twitter for iPhone  24173   5744   \n",
       "3  1022879441767288834 2018-07-27 16:20:37  Twitter for iPhone  35861   9474   \n",
       "4  1022877332858650624 2018-07-27 16:12:14  Twitter for iPhone  49944  12849   \n",
       "\n",
       "   SA  \n",
       "0   1  \n",
       "1   0  \n",
       "2   1  \n",
       "3   1  \n",
       "4   1  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "data.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [],
   "source": [
    "pos_tweet=data[data['SA']==1]['Tweet']\n",
    "neu_tweet=data[data['SA']==0]['Tweet']\n",
    "neg_tweet=data[data['SA']==-1]['Tweet']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "% of positive tweets =  50.0  %\n",
      "% of neutral tweets =  32.5  %\n",
      "% of negative tweets =  17.5  %\n"
     ]
    }
   ],
   "source": [
    "print (('% of positive tweets = '), (len(pos_tweet)*100)/len(data['Tweet']) , \" %\")\n",
    "print (('% of neutral tweets = '), (len(neu_tweet)*100)/len(data['Tweet']) , \" %\")\n",
    "print (('% of negative tweets = '), (len(neg_tweet)*100)/len(data['Tweet']) , \" %\")"
   ]
  },
  {
   "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": "3.6.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 1
}