project.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h1 style=\"color:blue;\">  <B> Tweets Clustering</B>  $~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~$<B> BY INES REBHI</B>\n",
    " </h1>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "\n",
    "<img src=\"images/giphy.gif\" width=\"400\" height=\"400\" align=\"center\"/>'\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3 style=\"color:blue;\"> <B>Introduction</B>\n",
    " </B> </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>Data redundancy is an important problem of Twitter. Twitter users are likely to generate similar tweets (e.g., using the Retweet function) about some popular topics/events.\n",
    "a result of a huge number of tweets which let  tweetos not interested to loss time about reading for the same topic many tweets     \n",
    "  So by clustering similar tweets together, we can generate a more concise and organized representation of the raw tweets, which will be very useful for busy Tweetos to read only one tweet per class  </B> </br>\n",
    "\n",
    "<B> Aim of our Project is to  </B>\n",
    "<li> <B>1/ gathering real time tweets  using a twitter API </B></li>\n",
    "<li><B> 2/ Preprocessing tweets </B> </li>\n",
    "<li><B> 3/Apply a NLP to text tweets</B> </li>\n",
    "<li><B> 4/ Modeling By using K-means as a ML algorithm of clustering </B></li>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3 style=\"color:blue;\"> <B>Keywords </B></h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>Text mining  $~~~/~~~~$clustering$~~~/~~~~$NLP $~~~/~~~~$tweepy$~~~/~~~~$ NLTK$~~~/~~~~$ twitter API </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3 style=\"color:blue;\"> <B>STEP1: Getting API keys from Twitter</B>\n",
    " </B> </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>An API is standing for Application Programming Interfaces (APIs) and they allow you to access resources only available on the server</B></br>\n",
    "<B>Now The Twitter API lets you read and write Twitter data. Thus, you can use it to compose tweets, read profiles, and access your followers’ data and a high volume of tweets on particular subjects in specific locations.</B></br>\n",
    "<B>So how can I get an API from twitter developer ??? </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<B style=\"color:orangered;\">1/ Create  twitter account </B>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/accounttwitter.PNG\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B style=\"color:orangered;\">2/Try this link to see how apply for a twitter api </B> </br></br>\n",
    "\n",
    "<B>https://www.youtube.com/watch?v=vlvtqp44xoQ </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B style=\"color:orangered;\">3/ Create a project in twitter developer</B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/Capture.PNG\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "</br>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B style=\"color:orangered;\">3/Get API  Keys </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/api.PNG\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B style=\"color:orangered;\"> 3/ Install Tweepy </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>Tweepy is an open source Python package that gives you a very convenient way to access the Twitter API with Python. Tweepy includes a set of classes and methods that represent Twitter’s models and API endpoints, and it transparently handles various implementation details, such as:</B>\n",
    "\n",
    "<B>* Rate limits</B>\n",
    "    \n",
    "<B>* Streams</B>\n",
    "\n",
    "<B> * Data encoding and decoding</B>\n",
    "    \n",
    "<B> * HTTP requests</B>\n",
    "    \n",
    "<B>* Results pagination</B>\n",
    "    \n",
    "<B>* OAuth authentication</B>\n",
    "    \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B style=\"color:orangered;\">4/ import Libraries  and credentials in our project </B> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "import tweepy           # To consume Twitter's API\n",
    "import pandas as pd     # To handle data\n",
    "import numpy as np      # For number computing\n",
    "\n",
    "# For plotting and visualization:\n",
    "from IPython.display import display\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "%matplotlib inline\n",
    "\n",
    "\n",
    "# Twitter App access keys for @user \n",
    "#credential part\n",
    "\n",
    "# Consume:\n",
    "CONSUMER_KEY    = 'pUruES5UQBACkbkcOolgfmjop'\n",
    "CONSUMER_SECRET = 'H0zY8T19bQIAMGKxTNEVZV4T52QNMfruGu8FLbECmizbw7qr7l'\n",
    "\n",
    "# Access:\n",
    "ACCESS_TOKEN  = '1324988873538228224-Sbn1EJXcxKeItAV33UzO7wX7TsFYjA'\n",
    "ACCESS_SECRET = 'MMyMliA8oNgEYspY3mKQ4Wi73F4YsD9CiELkyChQpxebo'\n",
    "auth = tweepy.OAuthHandler(CONSUMER_KEY, CONSUMER_SECRET)\n",
    "auth.set_access_token(ACCESS_TOKEN,ACCESS_SECRET)\n",
    "api=  tweepy.API(auth)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3><B style=\"color:blue;\">STEP2: Gathring DATA  </B> </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B style=\"color:orangered;\">1/ Live streaming tweets\n",
    "  </B> "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>In this Part we need to stream 10 000 tweets in real-time  from\n",
    "Twitter to solve the task of tweets clustering </B> "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<div class=\"alert alert-block alert-danger\">\n",
    "<b>Danger:</b> <B>Please don't run this cell because it can make you a lot of time to do the streaming ! </B> \n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "\n",
    "from tweepy import Stream\n",
    "from tweepy import OAuthHandler\n",
    "from tweepy.streaming import StreamListener\n",
    "import time\n",
    " \n",
    "class listener(StreamListener):\n",
    "    def on_data(self,data):\n",
    "       \n",
    "            saveFile = open('live_tweets.json','a')\n",
    "            saveFile.write(data)\n",
    "            saveFile.write('\\n')\n",
    "            saveFile.close()\n",
    "            return True\n",
    "       \n",
    "            time.sleep(5)\n",
    "    def on_status(self, status):\n",
    "        try:\n",
    "            if hasattr(status, 'retweeted_status') and hasattr(status.retweeted_status, 'extended_tweet'):\n",
    "                print('retweeted: ' + status.retweeted_status.extended_tweet['full_text'])\n",
    "            if hasattr(status, 'extended_tweet'):\n",
    "                print('extended_tweet: ' + status.extended_tweet['full_text'])\n",
    "            else:\n",
    "                print('text: ' + status.text)\n",
    "        except AttributeError:\n",
    "            print('attribute error: ' + status.text)\n",
    "auth = tweepy.OAuthHandler(CONSUMER_KEY, CONSUMER_SECRET)\n",
    "auth.set_access_token(ACCESS_TOKEN,ACCESS_SECRET)\n",
    "twitterStream = Stream(auth,listener(),tweet_mode='extended')\n",
    "\n",
    "twitterStream.filter(languages=[\"en\"],track=[\"politic\" , \"health\" , \"sport\",\"business\",\"entertaiment\"])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3><B style=\"color:blue;\">STEP3:DATA PREPARATION </B> </h3>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h3><B style=\"color:orangered;\">Data Processing and Wrangling  </B> </h3>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">><B style=\"color:green;\"> 1/ From json to csv </B> \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>the csv form  serve to make our dataset(tweets) more visualised than json form. So we need to change our dataset from json to csv </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<div class=\"info info-block alert-info\">\n",
    "<b>INFO:</b> <B>It's unused to run these two cells </B> \n",
    "</div>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 62,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "df = pd.read_json ('tweetsstreamed.json',lines = True)\n",
    "export_csv = df.to_csv ('bdtweets.csv', index = None, header=True)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 56,
   "metadata": {},
   "outputs": [],
   "source": [
    "df = pd.read_json ('live_tweets1.json',lines = True)\n",
    "export_csv = df.to_csv ('bdtweets2.csv', index = None, header=True)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">><B style=\"color:green;\"> 2/Cleaning text </B> \n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    ">><B>In this step we will \n",
    "<B><li>transform tweet text into lowercase</li></B>\n",
    "<B><li>remove twitter handles</li></B>\n",
    "<B><li>remove hyperlinks</li></B>\n",
    "    <B><li>remove non-alphanumeric characters such as punctuation marks </li></B>\n",
    "<B><li>remove whitespace</li></B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>a/keep only the text column and ignore the rest of features</B>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape of tweets dataframe with duplicated tweets (20422, 1)\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>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>@WTAJnews Great!  Another going out of busines...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>@DailyCaller @LisaMarieBoothe Health and educa...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>RT @MatthewJshow: @realDonaldTrump China is we...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Video: Demba Ba sees red as he stands up for h...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>RT @texman71: Fuck your restrictions!!</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text\n",
       "0  @WTAJnews Great!  Another going out of busines...\n",
       "1  @DailyCaller @LisaMarieBoothe Health and educa...\n",
       "2  RT @MatthewJshow: @realDonaldTrump China is we...\n",
       "3  Video: Demba Ba sees red as he stands up for h...\n",
       "4             RT @texman71: Fuck your restrictions!!"
      ]
     },
     "execution_count": 2,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd \n",
    "tweets = pd.read_csv(\"bdtweets2.csv\", usecols = ['text'])\n",
    "print(\"shape of tweets dataframe with duplicated tweets\",tweets.shape)\n",
    "tweets.head()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "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>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>20417</th>\n",
       "      <td>TAKE ACTION AND OPPOSE RUSHED APPROVALS WITHOU...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20418</th>\n",
       "      <td>Revive your aging skin and reduce the appearan...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20419</th>\n",
       "      <td>Busy 🐝!!!! \\n\\nWhen you own a business this is...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20420</th>\n",
       "      <td>RT @JeaneF1MSP: Dear @BBCScotNine, unpaid care...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20421</th>\n",
       "      <td>RT @hycfreedom: @HBO @HackedOffHugh HUNTER BID...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text\n",
       "20417  TAKE ACTION AND OPPOSE RUSHED APPROVALS WITHOU...\n",
       "20418  Revive your aging skin and reduce the appearan...\n",
       "20419  Busy 🐝!!!! \\n\\nWhen you own a business this is...\n",
       "20420  RT @JeaneF1MSP: Dear @BBCScotNine, unpaid care...\n",
       "20421  RT @hycfreedom: @HBO @HackedOffHugh HUNTER BID..."
      ]
     },
     "execution_count": 3,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets.tail()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "RT @SenTomCotton: This opinion reads like the delusional ramblings of a Resistance lawyer afflicted with Trump Derangement Syndrome. Judge…         186\n",
       "RT @btsportfootball: Nothing but respect ❤️\\n\\nMessi 🤝 Ronaldo\\n\\n📺 BT Sport ESPN HD https://t.co/Qo1KUQKZ7Q                                        171\n",
       "RT @SteveGuest: Joe Biden defeated AGAIN by the Teleprompter.\\n\\n\"For Secretary of Health and Education Services, I nominate Xavier Bacheria.…      153\n",
       "RT @MelissaReddy_: Demba Ba 👏🏽\\nIstanbul Basaksehir 👏🏽\\nPSG 👏🏽\\n\\nShowing us 'zero tolerance' is more than just empty words, can be easily acti…    137\n",
       "RT @SchmittNYC: Biden didn’t nominate Becerra - his handlers did                                                                                    120\n",
       "                                                                                                                                                   ... \n",
       "RT @jaskaran78: Definitely worth a read https://t.co/WvmP3nfDTb                                                                                       1\n",
       "@business I knew Michigan stop the OS winning streak this year.                                                                                       1\n",
       "Yass, mental health is so important                                                                                                                   1\n",
       "@andrewberreyesa I think the biggest benefit is I feel it’ll make us more competitive in and push the coaches to gr… https://t.co/ZhvWnCV256          1\n",
       "@ex0ticflwr It’s a cost benefit analysis. Malls provide more profit to the economy than a small family owned busine… https://t.co/X6bRCYzEOK          1\n",
       "Name: text, Length: 13937, dtype: int64"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets['text'].value_counts()\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><B>we are looking in our dataset and  note that we have many duplications that we must removing</B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>b/Remove duplicated tweets </B>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "shape of tweets dataframe After removing duplicated tweets (13937, 1)\n"
     ]
    }
   ],
   "source": [
    "\n",
    "tweets.drop_duplicates(subset='text',inplace=True)\n",
    "# tweets.drop_duplicates(subset=['text'], keep='first', inplace=True)  #tou can also run this line of code to drop duplicate tweets\n",
    "\n",
    "print(\"shape of tweets dataframe After removing duplicated tweets\",tweets.shape)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><B> we can showing top of hashtag in our dataset </B>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [],
   "source": [
    "import regex as re\n",
    "import numpy as np\n",
    "from collections import Counter\n",
    "\n",
    "tweets['hashtags'] = tweets['text'].apply(lambda twt : re.findall(r\"#(\\w+)\", twt))\n",
    "d = Counter(tweets.hashtags.sum())\n",
    "# print(d)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "3602"
      ]
     },
     "execution_count": 7,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets_hashtags = pd.DataFrame([d]).T\n",
    "tweets_hashtags.columns = ['freq']\n",
    "tweets_hashtags.freq.sum()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "top hashtag in our dataset\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>freq</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>COVID19</th>\n",
       "      <td>103</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>business</th>\n",
       "      <td>81</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>health</th>\n",
       "      <td>42</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>assignment</th>\n",
       "      <td>38</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>homework</th>\n",
       "      <td>33</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Nursing</th>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>research</th>\n",
       "      <td>31</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Literature</th>\n",
       "      <td>29</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>Essay</th>\n",
       "      <td>28</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>chemistry</th>\n",
       "      <td>26</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "            freq\n",
       "COVID19      103\n",
       "business      81\n",
       "health        42\n",
       "assignment    38\n",
       "homework      33\n",
       "Nursing       31\n",
       "research      31\n",
       "Literature    29\n",
       "Essay         28\n",
       "chemistry     26"
      ]
     },
     "execution_count": 8,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets_hashtags.sort_values(by=['freq'], ascending=False, inplace=True)\n",
    "print(\"top hashtag in our dataset\")\n",
    "tweets_hashtags.head(10)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Among 20000 Tweets, 3602 Hashtags were used.\n"
     ]
    },
    {
     "data": {
      "text/plain": [
       "Text(0.5, 1.0, 'Top 10 Hashtags of dataset')"
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 864x648 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "### Visualization of Top 10 Hashtags\n",
    "labels = tweets_hashtags.head(10).index.values.tolist()\n",
    "freq = tweets_hashtags['freq'].head(10).values.tolist()\n",
    "index = np.arange(len(freq))\n",
    "\n",
    "print(\"Among 20000 Tweets, 3602 Hashtags were used.\")\n",
    "plt.figure(figsize=(12,9))\n",
    "plt.bar(index, freq, alpha=0.8, color= 'orangered')\n",
    "plt.xlabel('Hashtags', fontsize=13)\n",
    "plt.ylabel('Frequency', fontsize=13)\n",
    "plt.xticks(index, labels, fontsize=11, rotation=90, fontweight=\"bold\") \n",
    "plt.title('Top 10 Hashtags of dataset', fontsize=12, fontweight=\"bold\",color='blue')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [],
   "source": [
    "tweets = tweets.drop(['hashtags'], axis=1)\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    " <B> c/extract the username through the tweets (preceded by @ or by RT @), emojis,punctuation</B></p> "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "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>text</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>great  another going out of business sale</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>health and education services is what he wan...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>china is well entrenched throughout business...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>video demba ba sees red as he stands up for hi...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>fuck your restrictions</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>...</th>\n",
       "      <td>...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20416</th>\n",
       "      <td>we haveright to education and good health we w...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20417</th>\n",
       "      <td>take action and oppose rushed approvals withou...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20418</th>\n",
       "      <td>revive your aging skin and reduce the appearan...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20419</th>\n",
       "      <td>busy  when you ownbusiness this is really the ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>20421</th>\n",
       "      <td>hunter biden lied tony bobulinski  the form...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>13937 rows × 1 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                    text\n",
       "0              great  another going out of business sale\n",
       "1        health and education services is what he wan...\n",
       "2        china is well entrenched throughout business...\n",
       "3      video demba ba sees red as he stands up for hi...\n",
       "4                                 fuck your restrictions\n",
       "...                                                  ...\n",
       "20416  we haveright to education and good health we w...\n",
       "20417  take action and oppose rushed approvals withou...\n",
       "20418  revive your aging skin and reduce the appearan...\n",
       "20419  busy  when you ownbusiness this is really the ...\n",
       "20421     hunter biden lied tony bobulinski  the form...\n",
       "\n",
       "[13937 rows x 1 columns]"
      ]
     },
     "execution_count": 11,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import regex as re\n",
    "\n",
    "\n",
    "#clean text\n",
    "\n",
    "def CleanText(TextToClean):\n",
    "    TextToClean.lower()  #Normalization\n",
    "    TextToClean = re.sub(r'@[A-Za-z0-9]+', '',str(TextToClean)) #to extract @\n",
    "    TextToClean= re.sub(r':','',str(TextToClean)) #to remove :\n",
    "    TextToClean= re.sub(r'\\n','',str(TextToClean)) #to remove  \\n\n",
    "    TextToClean= re.sub(r'#','',str(TextToClean)) #to remove #\n",
    "    TextToClean= re.sub(r'!','',str(TextToClean)) #to remove #\n",
    "    TextToClean = re.sub(r'RT[\\s]','',str(TextToClean)) #to remove #\n",
    "    TextToClean = re.sub(r'https','',str(TextToClean)) #to remove hyper links\n",
    "    TextToClean = re.sub(r'_','',str(TextToClean)) #to remove hyper links\n",
    "    TextToClean = re.sub(r'~~~@…','',str(TextToClean)) #to remove blanc space\n",
    "    TextToClean =re.sub(r\"\\&\\S*\\s\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\&\", \"\", str(TextToClean))\n",
    "    TextToClean= re.sub(r\"\\+\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\#\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\$\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\£\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\%\", \"\", str(TextToClean))\n",
    "    TextToClean= re.sub(r\"\\:\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\@\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\-\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\",\", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\" \\' \", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\"\\n \", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r'\\s+[a-zA-Z]\\s+', \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r\" \\? \", \"\", str(TextToClean))\n",
    "    TextToClean = re.sub(r'[^\\w\\s]','',str(TextToClean))\n",
    "    \n",
    "#     TextToClean = re.sub(r'\\s+', \"\", str(TextToClean))#Removing Extra Whitespaces\n",
    "#     TextToClean = re.sub(r'\\s+[a-zA-Z]\\s+', \"\", str(TextToClean))#Removing single carecter\n",
    "\n",
    "\n",
    "    return TextToClean\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "def remove_emoji(Text_with_emoji):\n",
    "    emoji_pattern = re.compile(\"[\"\n",
    "                               u\"\\U0001F600-\\U0001F64F\"  # emoticons\n",
    "                               u\"\\U0001F300-\\U0001F5FF\"  # symbols & pictographs\n",
    "                               u\"\\U0001F680-\\U0001F6FF\"  # transport & map symbols\n",
    "                               u\"\\U0001F1E0-\\U0001F1FF\"  # flags (appel phone)\n",
    "                               u\"\\U00002500-\\U00002BEF\"  # chinese char\n",
    "                               u\"\\U00002702-\\U000027B0\"\n",
    "                               u\"\\U00002702-\\U000027B0\"\n",
    "                               u\"\\U000024C2-\\U0001F251\"\n",
    "                               u\"\\U0001f926-\\U0001f937\"\n",
    "                               u\"\\U00010000-\\U0010ffff\"\n",
    "                               u\"\\u2640-\\u2642\"\n",
    "                               u\"\\u2600-\\u2B55\"\n",
    "                               u\"\\u200d\"\n",
    "                               u\"\\u23cf\"\n",
    "                               u\"\\u23e9\"\n",
    "                               u\"\\u231a\"\n",
    "                               u\"\\ufe0f\"  # dingbats\n",
    "                               u\"\\u3030\"\n",
    "                               \"]+\", flags=re.UNICODE)\n",
    "    return emoji_pattern.sub(r'', Text_with_emoji)\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "tweets['text'] = tweets['text'].apply(CleanText)\n",
    "tweets['text'] = tweets['text'].apply(remove_emoji)\n",
    "# tweets['extended_tweet'] = tweets['text'].apply(CleanText)\n",
    "# tweets['extended_tweet'] = tweets['text'].apply(remove_emoji)\n",
    "#lower text\n",
    "tweets[\"text\"] = tweets[\"text\"].str.lower()\n",
    "# tweets[\"extended_tweet\"] = tweets[\"text\"].str.lower()\n",
    "\n",
    "\n",
    "\n",
    "\n",
    "tweets"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "' new scathing analysis of s shady business by the  thereslot of questions around this guy the'"
      ]
     },
     "execution_count": 12,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets['text'][5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<div class=\"alert alert-block alert-success\">\n",
    "<b></b> <B> Now ! we have only cleaned and lower text </B>  \n",
    "\n",
    "</div>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3 style=\"color:blue;\"> <B>  Natural Language Processing </B>  </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>Natural language processing includes many different techniques for interpreting human language, ranging from statistical and machine learning methods to rules-based and algorithmic approaches.</B>\n",
    "</br>\n",
    "<B>Basic NLP tasks include tokenization and parsing, lemmatization/stemming, part-of-speech tagging, language detection and identification of semantic relationships.</B></br>\n",
    "<B>In general terms, NLP tasks break down language into shorter, elemental pieces, try to understand relationships between the pieces and explore how the pieces work together to create meaning.</B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<h4 style=\"color:orangered;\"> <B> Natural Language Toolkit (NLTK) for  analyzing text </B>  </h4>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>NLTK is a suite of libraries and programs for symbolic and statistical natural language processing (NLP) for English written in the Python programming language</B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Install NLTK </B></h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting nltk\n",
      "  Downloading nltk-3.5.zip (1.4 MB)\n",
      "Requirement already satisfied: click in c:\\users\\user\\anaconda3\\lib\\site-packages (from nltk) (7.1.2)\n",
      "Requirement already satisfied: joblib in c:\\users\\user\\anaconda3\\lib\\site-packages (from nltk) (0.16.0)\n",
      "Requirement already satisfied: regex in c:\\users\\user\\anaconda3\\lib\\site-packages (from nltk) (2020.6.8)\n",
      "Requirement already satisfied: tqdm in c:\\users\\user\\anaconda3\\lib\\site-packages (from nltk) (4.47.0)\n",
      "Building wheels for collected packages: nltk\n",
      "  Building wheel for nltk (setup.py): started\n",
      "  Building wheel for nltk (setup.py): finished with status 'done'\n",
      "  Created wheel for nltk: filename=nltk-3.5-py3-none-any.whl size=1434679 sha256=3bc961ddc00caade782f8044f76ddea6f35277e2fc0e79d914c76b880c08c97e\n",
      "  Stored in directory: c:\\users\\user\\appdata\\local\\pip\\cache\\wheels\\ff\\d5\\7b\\f1fb4e1e1603b2f01c2424dd60fbcc50c12ef918bafc44b155\n",
      "Successfully built nltk\n",
      "Installing collected packages: nltk\n",
      "Successfully installed nltk-3.5\n"
     ]
    }
   ],
   "source": [
    "!pip install nltk"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Install Conda package for NLTK </B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src =\"images/anacondaNLTK.PNG\" >"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B>Install punkt</B></h4>\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/punketinstalled.PNG\" >"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "><h4 style=\"color:orangered;\"><B> remove stopwords </B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "<B>In every language, some words are particularly common. While their use in the language is crucial, they don’t usually convey a particular meaning, especially if taken out of context. This is the case of articles, conjunctions, some adverbs, etc. which are commonly called stop-words.</br>\n",
    "Stop-word removal is one important step that should be considered during the pre-processing stages. </br>\n",
    "NLTK provides a simple list for English stop-words. </B>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/stopwords.PNG\" >"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Tokenization </B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>In this step , we will split tweets text into token (words) by using NLTK’s TweetTokenizer.It can  “collapse” repeated characters - that is, lolll, lollllll, and lollllllllllll will all collapse to the same representation \"lolll\" (three “l”s). This is helpful because we tend to think that these tokens represent approximately the same thing. This feature helps curb the curse of dimensionality (i.e. too many low-frequency tokens), while maintaining Twitter-specific features.  </B>\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "{'theirs', 'has', 'not', 'what', \"mightn't\", 'can', 'don', \"aren't\", \"haven't\", \"mustn't\", 'weren', 'had', 'before', 'nor', 'all', 'couldn', 'both', 'when', 'shouldn', 'those', \"it's\", 'am', 'at', 'ain', 'if', 'been', 'aren', 'out', 'only', 'rts', 'above', 'shan', 'herself', 'any', 'which', 'or', 'until', 'hers', 'and', 've', \"hadn't\", 'does', \"wouldn't\", 'was', 'over', \"hasn't\", 'again', 'do', 'himself', 'under', 'ma', 'is', 'most', 'off', 'but', 's', 'ourselves', 'll', 'with', \"you're\", 'haven', 'such', 'down', \"won't\", \"needn't\", 'hadn', 'it', 'each', 'the', 'up', \"weren't\", 'didn', 'i', 'yourself', \"shouldn't\", 'mightn', 'their', 'whom', 'for', 'by', 'be', 'because', 'his', 'so', 'then', 'yourselves', 'where', 'him', 'he', 'during', \"should've\", 'did', 'our', 'hasn', 'doing', 'here', 'few', 'too', \"she's\", 'rt', 'needn', 'are', 'as', 'to', 'itself', 'you', 'into', \"you'd\", 'mustn', 'doesn', 'myself', 'yours', 'a', 'how', \"doesn't\", \"wasn't\", 'more', 'your', 'o', 'who', 'an', 'won', 'should', \"you'll\", \"didn't\", 'wouldn', 'against', 'once', 'wasn', 't', 'were', 'between', 'm', 'no', 're', 'about', 'while', 'just', 'its', 'have', \"don't\", 'these', 'retweet', 'from', 'very', 'that', 'on', 'd', 'some', 'themselves', 'through', 'after', 'having', 'now', 'me', 'why', 'below', \"shan't\", 'will', 'further', 'in', 'them', 'own', 'other', \"couldn't\", 'of', 'y', \"you've\", 'same', 'we', 'she', \"that'll\", 'than', 'ours', 'my', 'there', 'they', 'her', 'isn', \"isn't\", 'this', 'being'}\n"
     ]
    }
   ],
   "source": [
    "#*************************************************************************************************\n",
    "                                               #This is the stopwords\n",
    "#*************************************************************************************************\n",
    "\n",
    "import nltk\n",
    "#nltk.download()\n",
    "from nltk.tokenize import TweetTokenizer\n",
    "from nltk.corpus import stopwords\n",
    "from nltk.tokenize import word_tokenize\n",
    "from nltk.corpus import stopwords\n",
    "\n",
    "additional  = ['rt','rts','retweet']\n",
    "swords = set().union(stopwords.words('english'),additional)\n",
    "print(swords)\n",
    "\n",
    "tweets['text'] = tweets['text'].apply(lambda x: ' '.join([i for i in x.split() if  i not in (swords)]))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "'new scathing analysis shady business thereslot questions around guy'"
      ]
     },
     "execution_count": 14,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets['text'][5]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [],
   "source": [
    "#*************************************************************************************************\n",
    "                                               #This is the tokenizer\n",
    "#*************************************************************************************************\n",
    "\n",
    "\n",
    "import nltk\n",
    "# nltk.download()\n",
    "from nltk.tokenize import TweetTokenizer\n",
    "from nltk.corpus import stopwords\n",
    "tweets['tokonized_text'] = '' \n",
    "\n",
    "tokenizer = TweetTokenizer()\n",
    "tweets['tokonized_text']  =tweets['text'].apply(tokenizer.tokenize)\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['new',\n",
       " 'scathing',\n",
       " 'analysis',\n",
       " 'shady',\n",
       " 'business',\n",
       " 'thereslot',\n",
       " 'questions',\n",
       " 'around',\n",
       " 'guy']"
      ]
     },
     "execution_count": 16,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets['tokonized_text'][5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Stemming </B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>Stemming is the process of reducing inflection in words to their root forms such as mapping a group of words to the same stem even if the stem itself is not a valid word in the Language.</B>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.stem import PorterStemmer\n",
    "ps = PorterStemmer()\n",
    "tweets['stemmed'] = tweets['tokonized_text'].apply(lambda x: [ps.stem(i) for i in x if i != ''])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['new',\n",
       " 'scath',\n",
       " 'analysi',\n",
       " 'shadi',\n",
       " 'busi',\n",
       " 'thereslot',\n",
       " 'question',\n",
       " 'around',\n",
       " 'guy']"
      ]
     },
     "execution_count": 19,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets['stemmed'][5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Lemmatization  </B></h4>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [],
   "source": [
    "from nltk.stem.wordnet import WordNetLemmatizer\n",
    "# nltk.download('wordnet') #download wordnet package with code \n",
    "lmtzr = WordNetLemmatizer()\n",
    "tweets['lemma'] = tweets['stemmed'].apply(lambda x: [lmtzr.lemmatize(word,'v') for word in x if word!= ''])\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "['new',\n",
       " 'scath',\n",
       " 'analysi',\n",
       " 'shadi',\n",
       " 'busi',\n",
       " 'thereslot',\n",
       " 'question',\n",
       " 'around',\n",
       " 'guy']"
      ]
     },
     "execution_count": 21,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "tweets['lemma'][5]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3 style=\"color:blue;\"> <B> STEP4 : Modeling </B>  </h3>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> K-means </B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>Now as a Ml Algorithm for clustering we will use K-means with cosine as a distance metric </B>\n",
    "\n",
    "<B>By this example we will understand How does K-means work </B>\n",
    "\n",
    "<img src=\"images/k-means.png\">\n",
    "<h5 style=\"color:tomato;\"><B> Step1</B></h5>\n",
    "Here we are having a few data points, which we want to cluster. So we would start by picking the number of clusters we want to have.</br>\n",
    "\n",
    "<h5 style=\"color:tomato;\"><B> Step2</B></h5>\n",
    "We have successfully marked the centers of these clusters. Now we will be marking all the points with respective colors on the basis of the distance they have from the centroid.</br>\n",
    "\n",
    "<h5 style=\"color:tomato;\"><B> Step3</B></h5>\n",
    "After marking all the data points, we will now be computing the centroid of this cluster again. We are doing it because initially, we had picked the centroid randomly. Then to remove error, if any, we are doing it.\n",
    "\n",
    "The centroid of the cluster is computed by finding a point within the cluster that would be equidistant from all the data points.</br>\n",
    "\n",
    "<h5 style=\"color:tomato;\"><B> Step4</B></h5>\n",
    "\n",
    "Now since we have computed the centroid again and we know it is not the same as it was before so we would iterate the process again and would find the points nearest to this centroid for each cluster.</br>\n",
    "\n",
    "<h5 style=\"color:tomato;\"><B> Step5</B></h5>\n",
    "Now we have got the result again. One may ask when shall we stop the iteration of this finding the centroid and then placing the data points accordingly? Well, you have to do it till the time when the position of the centroids doesn’t change.</br>\n",
    "\n",
    "<h5 style=\"color:tomato;\"><B> Step6</B></h5>\n",
    "We marked the two clusters.\n",
    "\n",
    "In this case, it was easy, so we were able to get the results in 2 iterations only.\n",
    "\n",
    "We had also talked about the random initialization that we are putting ourselves into. With this a problem we have is that it can land us up with some really bad clusters which won’t be of any use. "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> K-means Mertrics</B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B> K-means have 3 distance metrics </B>\n",
    "    <li><B>Euclidean Distance</B></li>\n",
    "    <li><B>Cosine Distance</B></li>\n",
    "    <li><B>Jaccard Similarity</B></li>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/metricc.gif\">"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [],
   "source": [
    "from sklearn.feature_extraction.text import TfidfVectorizer\n",
    "from sklearn.feature_extraction.text import CountVectorizer\n",
    "from sklearn.feature_extraction import text\n",
    "from nltk.cluster.kmeans import KMeansClusterer\n",
    "from sklearn.metrics.pairwise import cosine_similarity\n",
    "from sklearn.decomposition import TruncatedSVD\n",
    "from collections import Counter\n",
    "import matplotlib.pyplot as plt\n",
    "from datetime import datetime\n",
    "import nltk\n",
    "import pandas as pd\n",
    "import numpy as np\n",
    "import random\n",
    "from sklearn.decomposition import PCA      "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Vectorization</B></h4>\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>TFIDF is another way to convert textual data to numeric form, and is short for Term Frequency-Inverse Document Frequency. The vector value it yields is the product of these two terms; TF and IDF.\n",
    "Let’s first look at Term Frequency. We have already looked at term frequency with count vectorizer, but this time, we need one more step to calculate the relative frequency. Let’s say we have two documents in our corpus as below.</B>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "orig shape: (13937, 33694)\n",
      "new shape: (13937, 50)\n"
     ]
    }
   ],
   "source": [
    "RUNTIME_MODE = \"Cosine\"\n",
    "SAMPLE_COUNT = 14000\n",
    "MAX_CLUSTER = 25\n",
    "random.seed(1)\n",
    "\n",
    "content = tweets['text'].values # Get the news content to news array\n",
    "\n",
    "stop_words = set(text.ENGLISH_STOP_WORDS)\n",
    "my_stop_words = text.ENGLISH_STOP_WORDS.union([\n",
    " \"also\", \"said\", \"mr\", \"mrs\", \"im\", \"would\", \"could\", \"should\", \"first\", \"like\", \"dont\",\n",
    " \"wont\", \"get\", \"going\", \"ms\", \"one\", \"____\", \"_____\", \"new\", \"news\", \"told\", \"way\",\n",
    " \"year\", \"years\", \"don\", \"day\", \"man\", \"did\", \"just\", \"time\", \"times\", \"make\", \"000\",\n",
    " \"united\", \"state\", \"states\", \"people\", \"ve\", \"white\", \"house\", \"president\",\n",
    " \"government\", \"york\", \"want\", \"know\", \"think\", \"officials\", \"say\", \"breitbart\",\n",
    " \"percent\", \"home\", \"city\", \"case\", \"really\", \"work\", \"according\", \"including\",\n",
    " \"good\", \"campaign\", \"country\", \"long\", \"world\", \"donald\", \"trump\", \"didn\", \"women\",\n",
    " \"called\", \"american\", \"men\", \"later\", \"follow\", \"week\", \"black\", \"little\", \"company\",\n",
    " \"companies\", \"posted\", \"morning\", \"today\", \"evening\", \"com\", \"nytimes\", \"weekend\", \"10\",\"business\"\n",
    "]\n",
    ")\n",
    "vectorizer = CountVectorizer(stop_words=my_stop_words, binary=False)\n",
    "X_orig = vectorizer.fit_transform(content[0:SAMPLE_COUNT]).toarray()\n",
    "svd = TruncatedSVD(n_components=50, random_state = 0)\n",
    "X = svd.fit_transform(X_orig)\n",
    "print(\"orig shape:\", X_orig.shape)\n",
    "print(\"new shape:\", X.shape)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> K-means Clustering</B></h4>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\User\\anaconda3\\lib\\site-packages\\nltk\\cluster\\util.py:131: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  return 1 - (numpy.dot(u, v) / (sqrt(numpy.dot(u, u)) * sqrt(numpy.dot(v, v))))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Item count in clusters:\n",
      "Counter({1: 7707, 0: 6230})\n",
      "Item count in clusters:\n",
      "Counter({1: 7473, 0: 3365, 2: 3099})\n"
     ]
    }
   ],
   "source": [
    "terms = vectorizer.get_feature_names() # Get the array of terms that exists in documents\n",
    "inner_dist = np.zeros(shape=[MAX_CLUSTER - 1])\n",
    "for NUM_CLUSTERS in range(2, MAX_CLUSTER + 1):\n",
    "    kmeans = KMeansClusterer(num_means=NUM_CLUSTERS,\n",
    "    distance=nltk.cluster.util.cosine_distance, repeats=25, avoid_empty_clusters=True)\n",
    "#initial_means=means)\n",
    "    labels = kmeans.cluster(X, assign_clusters=True)\n",
    "    cluster_centers = np.asarray(kmeans.means())\n",
    "    print(\"Item count in clusters:\")\n",
    "    print(Counter(labels))\n",
    "    cluster_means = np.zeros(shape=[NUM_CLUSTERS, len(terms)])\n",
    "    item_counts = np.zeros(shape=[NUM_CLUSTERS])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\User\\anaconda3\\lib\\site-packages\\nltk\\cluster\\util.py:131: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  return 1 - (numpy.dot(u, v) / (sqrt(numpy.dot(u, u)) * sqrt(numpy.dot(v, v))))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Item count in clusters:\n",
      "Counter({0: 10802, 1: 3135})\n"
     ]
    }
   ],
   "source": [
    "terms = vectorizer.get_feature_names() # Get the array of terms that exists in documents\n",
    "NUM_CLUSTERS = 2\n",
    "inner_dist = np.zeros(shape=[MAX_CLUSTER - 1])\n",
    "kmeans = KMeansClusterer(num_means=NUM_CLUSTERS,distance=nltk.cluster.util.cosine_distance, repeats=25, avoid_empty_clusters=True)\n",
    "#initial_means=means)\n",
    "labels = kmeans.cluster(X, assign_clusters=True)\n",
    "cluster_centers = np.asarray(kmeans.means())\n",
    "print(\"Item count in clusters:\")\n",
    "print(Counter(labels))\n",
    "cluster_means = np.zeros(shape=[NUM_CLUSTERS, len(terms)])\n",
    "item_counts = np.zeros(shape=[NUM_CLUSTERS])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stderr",
     "output_type": "stream",
     "text": [
      "C:\\Users\\User\\anaconda3\\lib\\site-packages\\nltk\\cluster\\util.py:131: RuntimeWarning: invalid value encountered in double_scalars\n",
      "  return 1 - (numpy.dot(u, v) / (sqrt(numpy.dot(u, u)) * sqrt(numpy.dot(v, v))))\n"
     ]
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Item count in clusters:\n",
      "Counter({0: 7658, 3: 3000, 1: 2189, 2: 1090})\n"
     ]
    }
   ],
   "source": [
    "terms = vectorizer.get_feature_names() # Get the array of terms that exists in documents\n",
    "NUM_CLUSTERS = 4\n",
    "inner_dist = np.zeros(shape=[MAX_CLUSTER - 1])\n",
    "kmeans = KMeansClusterer(num_means=NUM_CLUSTERS,distance=nltk.cluster.util.cosine_distance, repeats=25, avoid_empty_clusters=True)\n",
    "#initial_means=means)\n",
    "labels = kmeans.cluster(X, assign_clusters=True)\n",
    "cluster_centers = np.asarray(kmeans.means())\n",
    "print(\"Item count in clusters:\")\n",
    "print(Counter(labels))\n",
    "cluster_means = np.zeros(shape=[NUM_CLUSTERS, len(terms)])\n",
    "item_counts = np.zeros(shape=[NUM_CLUSTERS])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "terms = vectorizer.get_feature_names() # Get the array of terms that exists in documents\n",
    "NUM_CLUSTERS = 6\n",
    "inner_dist = np.zeros(shape=[MAX_CLUSTER - 1])\n",
    "kmeans = KMeansClusterer(num_means=NUM_CLUSTERS,distance=nltk.cluster.util.cosine_distance, repeats=25, avoid_empty_clusters=True)\n",
    "#initial_means=means)\n",
    "labels = kmeans.cluster(X, assign_clusters=True)\n",
    "cluster_centers = np.asarray(kmeans.means())\n",
    "print(\"Item count in clusters:\")\n",
    "print(Counter(labels))\n",
    "cluster_means = np.zeros(shape=[NUM_CLUSTERS, len(terms)])\n",
    "item_counts = np.zeros(shape=[NUM_CLUSTERS])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "terms = vectorizer.get_feature_names() # Get the array of terms that exists in documents\n",
    "NUM_CLUSTERS = 10\n",
    "inner_dist = np.zeros(shape=[MAX_CLUSTER - 1])\n",
    "kmeans = KMeansClusterer(num_means=NUM_CLUSTERS,distance=nltk.cluster.util.cosine_distance, repeats=25, avoid_empty_clusters=True)\n",
    "#initial_means=means)\n",
    "labels = kmeans.cluster(X, assign_clusters=True)\n",
    "cluster_centers = np.asarray(kmeans.means())\n",
    "print(\"Item count in clusters:\")\n",
    "print(Counter(labels))\n",
    "cluster_means = np.zeros(shape=[NUM_CLUSTERS, len(terms)])\n",
    "item_counts = np.zeros(shape=[NUM_CLUSTERS])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B> Top words per Cluster</B></h4>\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top terms per cluster:\n",
      "Cluster 0 : sport  , \n",
      "small  , \n",
      "covid19  , \n",
      "need  , \n",
      "mind  , \n",
      "help  , \n",
      "Cluster 1 : health  , \n",
      "mental  , \n",
      "care  , \n",
      "public  , \n",
      "covid19  , \n",
      "team  , \n"
     ]
    }
   ],
   "source": [
    "SAMPLE_COUNT = 13500\n",
    "NUM_CLUSTERS = 2 \n",
    "\n",
    "for ind in range(SAMPLE_COUNT):\n",
    "    cluster_means[labels[ind]] = cluster_means[labels[ind]] + X_orig[ind]\n",
    "    item_counts[labels[ind]] = item_counts[labels[ind]] + 1\n",
    "for ind in range(NUM_CLUSTERS):\n",
    "    cluster_means[ind] = cluster_means[ind] / item_counts[ind]\n",
    "print(\"Top terms per cluster:\")\n",
    "order_centroids = cluster_means.argsort()[:, ::-1]\n",
    "\n",
    "for cls in range(NUM_CLUSTERS):\n",
    "    print(\"Cluster\", cls, ': ', end=''),\n",
    "    for ind in order_centroids[cls, :6]:\n",
    "        print(terms[ind], ' , ', end='')\n",
    "        print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 36,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Top terms per cluster:\n",
      "Cluster 0 : covid19  , \n",
      "Cluster 1 : small  , \n",
      "Cluster 2 : sport  , \n",
      "Cluster 3 : health  , \n"
     ]
    }
   ],
   "source": [
    "SAMPLE_COUNT = 13500\n",
    "NUM_CLUSTERS = 4\n",
    "\n",
    "for ind in range(SAMPLE_COUNT):\n",
    "    cluster_means[labels[ind]] = cluster_means[labels[ind]] + X_orig[ind]\n",
    "    item_counts[labels[ind]] = item_counts[labels[ind]] + 1\n",
    "for ind in range(NUM_CLUSTERS):\n",
    "    cluster_means[ind] = cluster_means[ind] / item_counts[ind]\n",
    "print(\"Top terms per cluster:\")\n",
    "order_centroids = cluster_means.argsort()[:, ::-1]\n",
    "\n",
    "for cls in range(NUM_CLUSTERS):\n",
    "    print(\"Cluster\", cls, ': ', end=''),\n",
    "    for ind in order_centroids[cls, :1]:\n",
    "        print(terms[ind], ' , ', end='')\n",
    "        print()\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "\n",
    "><h4 style=\"color:green;\"><B> PCA</B></h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "(13937, 2)"
      ]
     },
     "execution_count": 28,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dist = 1 - cosine_similarity(X)\n",
    "\n",
    "pca = PCA(2)\n",
    "pca.fit(dist)\n",
    "X_PCA = pca.transform(dist)\n",
    "X_PCA.shape"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B>Display Clusters  \n",
    "</B></h4>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 37,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x936 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "x, y = X_PCA[:, 0], X_PCA[:, 1]\n",
    "\n",
    "colors = {0: 'red',\n",
    "          1: 'blue',\n",
    "          2: 'yellow',\n",
    "          3:'green'\n",
    "         }\n",
    "\n",
    "names = { \n",
    "           0 :' covid19'  , \n",
    "           1 : 'small ' , \n",
    "            2 : 'sport ' , \n",
    "            3 : 'health ' \n",
    "         }\n",
    "  \n",
    "df = pd.DataFrame({'x': x, 'y':y, 'label':labels}) \n",
    "groups = df.groupby('label')\n",
    "\n",
    "fig, ax = plt.subplots(figsize=(20, 13)) \n",
    "\n",
    "for name, group in groups:\n",
    "    ax.plot(group.x, group.y, marker='o', linestyle='', ms=5,\n",
    "            color=colors[name],label=names[name], mec='none')\n",
    "    ax.set_aspect('auto')\n",
    "    ax.tick_params(axis='x',which='both',bottom='off',top='off',labelbottom='off')\n",
    "    ax.tick_params(axis= 'y',which='both',left='off',top='off',labelleft='off')\n",
    "    \n",
    "ax.legend()\n",
    "ax.set_title(\"Topics of streamed tweets\")\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "><h4 style=\"color:green;\"><B>Démo\n",
    "</B></h4>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<img src=\"images/démo.gif\">"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<h3 style=\"color:blue;\"> <B>Conclusion</B>\n",
    " </B> </h3>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>In the present project ,K-Means algorithm has been executed by</B>\n",
    "\n",
    "<li><B>data representation method :TF-IDF</B></li>\n",
    "<li><B>Distance metrics : Cosine Similarity</B></li>\n",
    "<li><B>k =6 values (2 to 6 clusters)</B></li>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B> References </B>\n",
    "\n",
    "https://www.datacamp.com/community/tutorials/stemming-lemmatization-python</br>\n",
    "https://datascienceplus.com/twitter-analysis-with-python/</br>\n",
    "https://berkeley-stat159-f17.github.io/stat159-f17/lectures/11-strings/11-nltk..html</br>\n",
    "https://www.xspdf.com/resolution/52940753.html</br>\n",
    "https://medium.com/analytics-vidhya/exploring-twitter-data-using-python-af1287ee65f1</br>\n",
    "https://towardsdatascience.com/kmeans-clustering-for-classification-74b992405d0a\n",
    "https://www.xspdf.com/resolution/55501459.html"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<B>For requirements.txt</B>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Collecting watermark\n",
      "  Downloading watermark-2.1.0-py2.py3-none-any.whl (5.7 kB)\n",
      "Requirement already satisfied: ipython in c:\\users\\user\\anaconda3\\lib\\site-packages (from watermark) (7.16.1)\n",
      "Requirement already satisfied: colorama; sys_platform == \"win32\" in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (0.4.3)\n",
      "Requirement already satisfied: backcall in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (0.2.0)\n",
      "Requirement already satisfied: prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (3.0.5)\n",
      "Requirement already satisfied: decorator in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (4.4.2)\n",
      "Requirement already satisfied: jedi>=0.10 in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (0.17.1)\n",
      "Requirement already satisfied: pygments in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (2.6.1)\n",
      "Requirement already satisfied: traitlets>=4.2 in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (4.3.3)\n",
      "Requirement already satisfied: pickleshare in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (0.7.5)\n",
      "Requirement already satisfied: setuptools>=18.5 in c:\\users\\user\\anaconda3\\lib\\site-packages (from ipython->watermark) (49.2.0.post20200714)\n",
      "Requirement already satisfied: wcwidth in c:\\users\\user\\anaconda3\\lib\\site-packages (from prompt-toolkit!=3.0.0,!=3.0.1,<3.1.0,>=2.0.0->ipython->watermark) (0.2.5)\n",
      "Requirement already satisfied: parso<0.8.0,>=0.7.0 in c:\\users\\user\\anaconda3\\lib\\site-packages (from jedi>=0.10->ipython->watermark) (0.7.0)\n",
      "Requirement already satisfied: six in c:\\users\\user\\anaconda3\\lib\\site-packages (from traitlets>=4.2->ipython->watermark) (1.15.0)\n",
      "Requirement already satisfied: ipython-genutils in c:\\users\\user\\anaconda3\\lib\\site-packages (from traitlets>=4.2->ipython->watermark) (0.2.0)\n",
      "Installing collected packages: watermark\n",
      "Successfully installed watermark-2.1.0\n",
      "Note: you may need to restart the kernel to use updated packages.\n"
     ]
    }
   ],
   "source": [
    "pip install watermark\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 40,
   "metadata": {},
   "outputs": [],
   "source": [
    "%load_ext watermark\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "\u001b[1;31mDocstring:\u001b[0m\n",
       "::\n",
       "\n",
       "  %watermark [-a AUTHOR] [-d] [-n] [-t] [-i] [-z] [-u] [-c CUSTOM_TIME] [-v] [-p PACKAGES] [-h] [-m] [-g] [-r]\n",
       "                 [-b] [-w] [-iv]\n",
       "\n",
       "IPython magic function to print date/time stamps\n",
       "and various system information.\n",
       "\n",
       "optional arguments:\n",
       "  -a AUTHOR, --author AUTHOR\n",
       "                        prints author name\n",
       "  -d, --date            prints current date as YYYY-mm-dd\n",
       "  -n, --datename        prints date with abbrv. day and month names\n",
       "  -t, --time            prints current time as HH-MM-SS\n",
       "  -i, --iso8601         prints the combined date and time including the time zone in the ISO 8601 standard with UTC\n",
       "                        offset\n",
       "  -z, --timezone        appends the local time zone\n",
       "  -u, --updated         appends a string \"Last updated: \"\n",
       "  -c CUSTOM_TIME, --custom_time CUSTOM_TIME\n",
       "                        prints a valid strftime() string\n",
       "  -v, --python          prints Python and IPython version\n",
       "  -p PACKAGES, --packages PACKAGES\n",
       "                        prints versions of specified Python modules and packages\n",
       "  -h, --hostname        prints the host name\n",
       "  -m, --machine         prints system and machine info\n",
       "  -g, --githash         prints current Git commit hash\n",
       "  -r, --gitrepo         prints current Git remote address\n",
       "  -b, --gitbranch       prints current Git branch\n",
       "  -w, --watermark       prints the current version of watermark\n",
       "  -iv, --iversions      prints the name/version of all imported modules\n",
       "\u001b[1;31mFile:\u001b[0m      c:\\users\\user\\anaconda3\\lib\\site-packages\\watermark\\watermark.py\n"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "%watermark?\n",
    "\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 41,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "numpy     : 1.18.5\n",
      "matplotlib: 3.2.2\n",
      "pandas    : 1.0.5\n",
      "regex     : 2020.6.8\n",
      "nltk      : 3.5\n",
      "sklearn   : 0.22\n",
      "seaborn   : 0.10.1\n",
      "tweepy    : 3.9.0\n",
      "\n"
     ]
    }
   ],
   "source": [
    "%watermark --iversions"
   ]
  }
 ],
 "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": "3.8.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}