DataMining_twitterXmongodb.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "# <p><font color=deepskyblue>&#9658; INTRODUCTION TO SCRAPING AND DATA MINING &#9664;</font></p>\n",
    "# <p><font color=deepskyblue>Collect Tweets's Data into MongoDB and visualise them with NetworkX </font></p>\n",
    "\n",
    "\n",
    "___\n",
    "# <p><font color=teal> &#129414; Objectives </font></p>\n",
    "\n",
    "**We want to be able to extract tweets based on one or several specific hastags from twitter to then analyse them.**\n",
    "\n",
    "To achieve this goal we will go through these steps:\n",
    "\n",
    "- **[Set Up](#setup)** (1h)\n",
    "    - We assume that you already have Jupyter Notebook and are looking at it right now\n",
    "    - Install **MongoDB**\n",
    "    - Get your personal credentals from the **Twitter API**\n",
    "    - Select some **hastags** related to an event or movement (/!\\ we will use tweepy stream and you cannot collect tweets older than one week with it, so the event has to be in trends)\n",
    "- **[Data Mining](#datamining)** (1h)\n",
    "    - Use the Stream API to **collect tweets** to create our dataset (I collected 5000 tweets in another project, I guess 1000 would be a minimum, so it's better when it's a very trendy topic like a political election or an international sports event for instance)\n",
    "- **[Data Analysis](#dataanalysis)** (1h)\n",
    "    - Analyse the information we collected\n",
    "    - **Visualise** networks with graphs\n",
    "\n",
    "___\n",
    "# <p><font color=teal> &#129414; Useful sources </font></p>\n",
    "\n",
    "* [**Mining the Social Web, 2nd Edition by Matthew A. Russell**](https://www.webpages.uidaho.edu/~stevel/504/Mining-the-Social-Web-2nd-Edition.pdf) (really cool and detailed book for mining Twitter, Facebook, Linkedin, Google+, web pages, etc!)\n",
    "* [twitter doc](https://developer.twitter.com/en/docs/tweets/filter-realtime/api-reference/post-statuses-filter) the API limits permit you to do the requests inside the boundaries of the API\n",
    "* composition of a [tweet object](https://developer.twitter.com/en/docs/tweets/data-dictionary/overview/tweet-object)\n",
    "* [Stevesie Data youtube channel](https://www.youtube.com/c/StevesieData/videos) (Learn how to scrape data at scale from any public API)\n",
    "\n",
    "\n",
    "___\n",
    "# <p><font color=teal> &#129414; Frequent questions </font></p>\n",
    "\n",
    "* **What is an API ?** <br> An Application Programming Interface (API) is a software intermediary that allows two applications to talk to each other. In other words, an API is the _messenger_ that delivers your request to the provider that you're requesting it from and then delivers the response back to you. It is a tool set that programmers can use in helping them create software.\n",
    "* **What scrapping / data mining ?** <br> _Scraping_ refers to the process of extracting data from web sources and structuring it into a more convenient format. It does not involve any data processing or analysis. (you _collect_ data) <br>_Data mining_ is defined as a process used to extract usable data from a larger set of any raw data. It require knowledge in computer science, databases and statistics. Data mining refers to the process of analyzing large datasets to uncover trends and valuable insights. It does not involve any data gathering or extraction. You _look_ at the (most of the time HUGE) dataset that you have created by extracting data from somewhere. (you _organise_ data) \n",
    "* **What is MongoDB ?**<br> MongoDB is a _NoSQL database_ (it means that the data isn't stored in tables but in documents. It is useful for projects that need to be developed rapidely when you don't have a full overview and it offers flexibility, so it is less strict).<br><br>\n",
    "* **Is it legal ?** <br> Yes, the application can only access public information on Twitter. You can check out twitter's rules and policies about their API usage [here](https://help.twitter.com/en/rules-and-policies/twitter-api), and you can watch the video [Is Data Scraping Legal in 2020? Recent Court Rulings & Data Scraping Guidelines For Staying Legal](https://youtu.be/btjtuMsfdb4) by Stevesie Data.\n",
    "* **Is it free ?** <br> Twitter Streaming API is free to use but gives you limited results (and limited licensing usage of the data)\n",
    "* **Do I need to have a twitter account or give my data to twitter ?** <br> Yes, to have access to the API, you need to activate your \"developer account\", you will provide information about how you will use the collected data and give your phone number. Otherwise you cannot have access to the stream of tweets. If you don't want to do it, you still can use the json file of collected tweets that I provided in the repository. An other solution is to download .har files from a browser<br><br>\n",
    "* **What is the limit of tweets that you can collect ?** <br> For the GET method (the one that we use to get the tweets) the limit is set to 100,000 requests per day.\n",
    "* **How old are the tweets that I can collect ? Can I collect tweets from the FIFA world cup 2010 ?** <br> The Twitter Search API searches against a _sampling_ of recent Tweets published in the _past 7 days_. With the Premium or Entreprise versions of the API, you can have access to tweets from the past 30 days, or have access to archives back to 2006. However there is something called \"[Sandbox](https://developer.twitter.com/en/docs/twitter-api/premium/search-api/overview)\" (we don't use it here) that gives you access to archive data with a limited amount of requests. Another trick can be using .har files\n",
    "* **What kind of data do you have access to ?** <br> You have access to public twitter information: names, number of likes, text of tweets, number of followers, language, etc. Here is a [tweet object](https://developer.twitter.com/en/docs/twitter-api/v1/data-dictionary/object-model/tweet) that can give you an idea of what you can get.\n",
    "\n",
    "* **Does it work only for hashtags ?** <br> No, with words as well, but hastags assure a standardized formatting for your request\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "___\n",
    "# <p><font color=teal> &#129414; Set Up </font></p> <a id=\"setup\"></a>\n",
    "\n",
    "\n",
    "* Install [pymongo](https://pymongo.readthedocs.io/en/stable/installation.html) `python -m pip install pymongo` ([installation youtube video for Windows (no command line)](https://youtu.be/FwMwO8pXfq0))\n",
    "* Install [MongoDB Compass](https://www.mongodb.com/try/download/compass)\n",
    "* Install [tweepy](http://docs.tweepy.org/en/latest/install.html) `pip install tweepy`\n",
    "* [Create a twitter application](https://developer.twitter.com/en/application/intent) on [Twitter API](https://apps.twitter.com/) and get your acess tokens / credentials <br>\n",
    "(TUTORIAL FOR THIS STEP: [Authenticate a Python Application with Twitter using Tweepy](https://www.digitalocean.com/community/tutorials/how-to-authenticate-a-python-application-with-twitter-using-tweepy-on-ubuntu-14-04)). You will need to enter personal data like your phone number and get validation to have access to developer applications\n",
    "* Install [networkx](https://networkx.org/) and [pyvis](https://pyvis.readthedocs.io/en/latest/install.html) for the visualisation part `pip install networkx` and `pip install pyvis`\n",
    "\n",
    "<p><font color=white>https://rso6eziprue.typeform.com/to/i11T93mb</font></p>"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <p><font color=lightsteelblue> &#129370; PART 1: Collect the tweets </font></p> <a id=\"datamining\"></a>\n",
    "### a) Import libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "# access the twitter API\n",
    "from tweepy.streaming import StreamListener\n",
    "from tweepy import OAuthHandler\n",
    "from tweepy import Stream\n",
    "# work with json files\n",
    "import json\n",
    "# database mongodb working with python\n",
    "import pymongo\n",
    "# mongo connection\n",
    "from pymongo import MongoClient"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### b) Load your personal secret keys\n",
    "\n",
    "You need to create a `.keys.json` file in the root of your folder with your personal access tokens that you got on the [Twitter API](https://apps.twitter.com/) as explained in Set Up:\n",
    "\n",
    "`[{\n",
    "    \"consumer_key\": \"xxxxx\",\n",
    "    \"consumer_secret\": \"xxxxx\",\n",
    "    \"access_token\": \"\",\n",
    "    \"access_token_secret\": \"xxxxx\"\n",
    "}]`\n",
    "\n",
    "If you don't want to use Twitter API, you won't be able to do the next steps of collecting the tweets. You will be able to use the .json file that I provided and that will be used in **PART2 b**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Open and load the credentals stored in the file .keys.json\n",
    "with open('.keys.json', 'r') as myFile:\n",
    "    data = json.load(myFile)\n",
    "\n",
    "consumer_key = data[0]['consumer_key']\n",
    "consumer_secret = data[0]['consumer_secret']\n",
    "access_token = data[0]['access_token']\n",
    "access_token_secret = data[0]['access_token_secret']"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### c) Fetch the stream of tweets\n",
    "\n",
    "**Twitter time !** Now you need to choose what you want to collect so go in the top tweets and see what's up !\n",
    "\n",
    "Here we will use **MongoDB** with tweepy ! What is MongoDB ? <br>\n",
    "MongoDB is a **NoSQL database** (it means that the data isn't stored in tables but in documents. It is useful for projects that need to be developed rapidely when you don't have a full overview and it offers flexibility, so it is less strict). <br>\n",
    "The **documents** are saved into **JSON format**.<br>\n",
    "**Databases** hold one or more collections of documents.<br>\n",
    "MongoDB stores documents in **collections**. Collections are analogous to tables in relational databases.<br>\n",
    "(Learn more in [Mongo DB documentation](https://docs.mongodb.com/manual/core/databases-and-collections/))\n",
    "\n",
    "<img src=\"MongoDBdbcol.png\" align=\"left\" style=\"width: 500px;\"/>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "START COLLECTING TWEETS\n",
      "<class 'str'>\n",
      "1\n",
      "<class 'str'>\n",
      "2\n",
      "--- STOP THE SCRIPT ---\n"
     ]
    }
   ],
   "source": [
    "# Tweet collect\n",
    "client = MongoClient('localhost', 27017)\n",
    "db = client['analysis'] # your database name // example: client['YOUR_DB_NAME']\n",
    "collection = db['loisecuriteglobale3'] # your collection name // example: db['YOUR_COLLECTION_NAME']\n",
    "count_collectedtweets = 0 # made to keep track of how much tweets you have collected\n",
    "\n",
    "class StdOutListener(StreamListener):\n",
    "    \"\"\" A listener handles tweets that are received from the stream.\n",
    "    This is a basic listener that just prints received tweets to stdout.\n",
    "    \"\"\"\n",
    "\n",
    "    def on_data(self, data):\n",
    "        print(type(data))\n",
    "        collection.insert_one(json.loads(data)) # add the collected tweet into the collection\n",
    "        global count_collectedtweets \n",
    "        count_collectedtweets = count_collectedtweets+1\n",
    "        print(count_collectedtweets) # print the number of the collected tweet while collecting\n",
    "        # Fix a limit of tweets to collect (I would say that at least 1000 would be nice)\n",
    "        if count_collectedtweets < 100:\n",
    "            return True #collect\n",
    "        else:\n",
    "            return False #stop\n",
    "\n",
    "    def on_error(self, status):\n",
    "        print(status)\n",
    "\n",
    "if __name__ == '__main__':\n",
    "    l = StdOutListener()\n",
    "    auth = OAuthHandler(consumer_key, consumer_secret)\n",
    "    auth.set_access_token(access_token, access_token_secret)\n",
    "\n",
    "    stream = Stream(auth, l)\n",
    "    print('START COLLECTING TWEETS')\n",
    "    stream.filter(track=['loisecuriteglobale', 'violencepoliciere','racisme'])\n",
    "    print('--- STOP THE SCRIPT ---')\n",
    "    #example: stream.filter(track=['GJ', 'giletsjaunes', 'greve','macron','retraites']) #results in french\n",
    "    #example: stream.filter(track=['covid19', 'covid', 'corona','confinement','masques']) #results in various languages"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "<p><font color=white>\n",
    "# alternative fast way\n",
    "#auth = tweepy.OAuthHandler(consumer_key, consumer_secret)\n",
    "#auth.set_access_token(access_token, access_token_secret)\n",
    "#api = tweepy.API(auth)\n",
    "\n",
    "#results = api.search(q='cheese', count=100)\n",
    "\n",
    "#json_data = [r._json for r in results]\n",
    "\n",
    "#df = pd.io.json.json_normalize(json_data)\n",
    "</font></p> "
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <p><font color=lightsteelblue> &#128035; PART 2: Create a database </font></p>\n",
    "\n",
    "### a) Check and import the collection inside a .json file"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "If you use MongoDB compass, connect locally, then export your .json file and save it in your root folder\n",
    "\n",
    "Connection string to connect locally: **mongodb://localhost:27017/admin?readPreference=primary&appname=MongoDB%20Compass&ssl=false**\n",
    "\n",
    "\n",
    "<img src=\"MongoDBcompassLocalhost.png\" style=\"width: 500px;\"/>\n",
    "<img src=\"exportcollectioninjson.png\" style=\"width: 500px;\"/>\n",
    "<img src=\"exportcollectioninjson2.png\" style=\"width: 500px;\"/><br>\n",
    "\n",
    "Otherwise you can use the MongoDB prompt:\n",
    "\n",
    "`mongoexport --db YOUR_DB_NAME --collection YOUR_COLLECTION_NAME --out YOUR_COLLECTION_NAME.json`\n",
    "\n",
    "example:`mongoexport --db analysis --collection loisecuriteglobale3 --out loisecuriteglobale3.json`\n",
    "\n",
    "If you already have a collection, you can import it into MongoDB with:\n",
    "\n",
    "`mongoimport -d YOUR_DB_NAME -c brexit  --file YOUR_COLLECTION_NAME.json`"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### b) Display your file\n",
    "\n",
    "Once you saved the .json file in your root folder, you can check it quickly with this:"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>_id</th>\n",
       "      <th>contributors</th>\n",
       "      <th>coordinates</th>\n",
       "      <th>created_at</th>\n",
       "      <th>display_text_range</th>\n",
       "      <th>entities</th>\n",
       "      <th>extended_entities</th>\n",
       "      <th>extended_tweet</th>\n",
       "      <th>favorite_count</th>\n",
       "      <th>favorited</th>\n",
       "      <th>...</th>\n",
       "      <th>quoted_status_permalink</th>\n",
       "      <th>reply_count</th>\n",
       "      <th>retweet_count</th>\n",
       "      <th>retweeted</th>\n",
       "      <th>retweeted_status</th>\n",
       "      <th>source</th>\n",
       "      <th>text</th>\n",
       "      <th>timestamp_ms</th>\n",
       "      <th>truncated</th>\n",
       "      <th>user</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>{'$oid': '5fd8fb03c807ac4dda964830'}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020-12-15 18:05:49</td>\n",
       "      <td>[{'$numberInt': '38'}, {'$numberInt': '140'}]</td>\n",
       "      <td>{'hashtags': [], 'urls': [{'url': 'https://t.c...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'full_text': '@siironique @Raheem0971 @twm1s ...</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>False</td>\n",
       "      <td>NaN</td>\n",
       "      <td>&lt;a href=\"https://mobile.twitter.com\" rel=\"nofo...</td>\n",
       "      <td>@siironique @Raheem0971 @twm1s @lzylv \"il faut...</td>\n",
       "      <td>2020-12-15 18:05:49.571</td>\n",
       "      <td>True</td>\n",
       "      <td>{'id': {'$numberLong': '1281669801249775617'},...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>{'$oid': '5fd8fb07c807ac4dda964831'}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020-12-15 18:05:53</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'hashtags': [], 'urls': [], 'user_mentions': ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>{'url': 'https://t.co/UvLXEm0pTl', 'expanded':...</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>False</td>\n",
       "      <td>{'created_at': 'Tue Dec 15 17:57:05 +0000 2020...</td>\n",
       "      <td>&lt;a href=\"https://mobile.twitter.com\" rel=\"nofo...</td>\n",
       "      <td>RT @Nantes_Revoltee: Nantes ce soir : moment h...</td>\n",
       "      <td>2020-12-15 18:05:53.562</td>\n",
       "      <td>False</td>\n",
       "      <td>{'id': {'$numberInt': '271439132'}, 'id_str': ...</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>{'$oid': '5fd8fb09c807ac4dda964832'}</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>2020-12-15 18:05:55</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'hashtags': [], 'urls': [], 'user_mentions': ...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>False</td>\n",
       "      <td>...</td>\n",
       "      <td>NaN</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>{'$numberInt': '0'}</td>\n",
       "      <td>False</td>\n",
       "      <td>{'created_at': 'Tue Dec 15 15:45:25 +0000 2020...</td>\n",
       "      <td>&lt;a href=\"http://twitter.com/download/android\" ...</td>\n",
       "      <td>RT @sebastianjroche: Devant la dÃ©nÃ©gation du...</td>\n",
       "      <td>2020-12-15 18:05:55.318</td>\n",
       "      <td>False</td>\n",
       "      <td>{'id': {'$numberLong': '717476046900166657'}, ...</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "<p>3 rows × 37 columns</p>\n",
       "</div>"
      ],
      "text/plain": [
       "                                    _id  contributors  coordinates  \\\n",
       "0  {'$oid': '5fd8fb03c807ac4dda964830'}           NaN          NaN   \n",
       "1  {'$oid': '5fd8fb07c807ac4dda964831'}           NaN          NaN   \n",
       "2  {'$oid': '5fd8fb09c807ac4dda964832'}           NaN          NaN   \n",
       "\n",
       "           created_at                             display_text_range  \\\n",
       "0 2020-12-15 18:05:49  [{'$numberInt': '38'}, {'$numberInt': '140'}]   \n",
       "1 2020-12-15 18:05:53                                            NaN   \n",
       "2 2020-12-15 18:05:55                                            NaN   \n",
       "\n",
       "                                            entities extended_entities  \\\n",
       "0  {'hashtags': [], 'urls': [{'url': 'https://t.c...               NaN   \n",
       "1  {'hashtags': [], 'urls': [], 'user_mentions': ...               NaN   \n",
       "2  {'hashtags': [], 'urls': [], 'user_mentions': ...               NaN   \n",
       "\n",
       "                                      extended_tweet       favorite_count  \\\n",
       "0  {'full_text': '@siironique @Raheem0971 @twm1s ...  {'$numberInt': '0'}   \n",
       "1                                                NaN  {'$numberInt': '0'}   \n",
       "2                                                NaN  {'$numberInt': '0'}   \n",
       "\n",
       "   favorited                        ...                          \\\n",
       "0      False                        ...                           \n",
       "1      False                        ...                           \n",
       "2      False                        ...                           \n",
       "\n",
       "                             quoted_status_permalink          reply_count  \\\n",
       "0                                                NaN  {'$numberInt': '0'}   \n",
       "1  {'url': 'https://t.co/UvLXEm0pTl', 'expanded':...  {'$numberInt': '0'}   \n",
       "2                                                NaN  {'$numberInt': '0'}   \n",
       "\n",
       "         retweet_count  retweeted  \\\n",
       "0  {'$numberInt': '0'}      False   \n",
       "1  {'$numberInt': '0'}      False   \n",
       "2  {'$numberInt': '0'}      False   \n",
       "\n",
       "                                    retweeted_status  \\\n",
       "0                                                NaN   \n",
       "1  {'created_at': 'Tue Dec 15 17:57:05 +0000 2020...   \n",
       "2  {'created_at': 'Tue Dec 15 15:45:25 +0000 2020...   \n",
       "\n",
       "                                              source  \\\n",
       "0  <a href=\"https://mobile.twitter.com\" rel=\"nofo...   \n",
       "1  <a href=\"https://mobile.twitter.com\" rel=\"nofo...   \n",
       "2  <a href=\"http://twitter.com/download/android\" ...   \n",
       "\n",
       "                                                text            timestamp_ms  \\\n",
       "0  @siironique @Raheem0971 @twm1s @lzylv \"il faut... 2020-12-15 18:05:49.571   \n",
       "1  RT @Nantes_Revoltee: Nantes ce soir : moment h... 2020-12-15 18:05:53.562   \n",
       "2  RT @sebastianjroche: Devant la dÃ©nÃ©gation du... 2020-12-15 18:05:55.318   \n",
       "\n",
       "   truncated                                               user  \n",
       "0       True  {'id': {'$numberLong': '1281669801249775617'},...  \n",
       "1      False  {'id': {'$numberInt': '271439132'}, 'id_str': ...  \n",
       "2      False  {'id': {'$numberLong': '717476046900166657'}, ...  \n",
       "\n",
       "[3 rows x 37 columns]"
      ]
     },
     "execution_count": 4,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import pandas as pd\n",
    "data = pd.read_json('loisecuriteglobale3.json', lines=True)\n",
    "data.head(3)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### c) Format dataframe in a nice way\n",
    "(I reused the code of [hectoramirez](https://github.com/hectoramirez/Language-localization_FIFA) for this part, feel free to modify it to obtain a different formating)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [],
   "source": [
    "import glob\n",
    "\n",
    "# list all files containing tweets\n",
    "files = list(glob.iglob('loisecuriteglobale3.json'))\n",
    "\n",
    "tweets_data = []\n",
    "for file in files:\n",
    "    \n",
    "    tweets_file = open(file, \"r\", encoding = 'utf-8')\n",
    "\n",
    "    # Read in tweets and store in list: tweets_data\n",
    "    for line in tweets_file:\n",
    "        tweet = json.loads(line)\n",
    "        tweets_data.append(tweet)\n",
    "\n",
    "    tweets_file.close()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 100 tweets in the dataset.\n"
     ]
    }
   ],
   "source": [
    "print('There are', len(tweets_data), 'tweets in the dataset.')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [],
   "source": [
    "def flatten_tweets(tweets):\n",
    "    \"\"\" Flattens out tweet dictionaries so relevant JSON is \n",
    "        in a top-level dictionary. \"\"\"\n",
    "    \n",
    "    tweets_list = []\n",
    "    \n",
    "    # Iterate through each tweet\n",
    "    for tweet_obj in tweets:\n",
    "    \n",
    "        ''' User info'''\n",
    "        # Store the user screen name in 'user-screen_name'\n",
    "        tweet_obj['user-screen_name'] = tweet_obj['user']['screen_name']\n",
    "        \n",
    "        # Store the user screen name in 'user-name-id'\n",
    "        tweet_obj['user-name-id'] = tweet_obj['user']['id_str']\n",
    "        \n",
    "        # Store the user location\n",
    "        tweet_obj['user-location'] = tweet_obj['user']['location']\n",
    "    \n",
    "        ''' Text info'''\n",
    "        # Check if this is a 140+ character tweet\n",
    "        if 'extended_tweet' in tweet_obj:\n",
    "            # Store the extended tweet text in 'extended_tweet-full_text'\n",
    "            tweet_obj['extended_tweet-full_text'] = \\\n",
    "                                    tweet_obj['extended_tweet']['full_text']\n",
    "    \n",
    "        if 'retweeted_status' in tweet_obj:\n",
    "            # Store the retweet user screen name in \n",
    "            # 'retweeted_status-user-screen_name'\n",
    "            tweet_obj['retweeted_status-user-screen_name'] = \\\n",
    "                        tweet_obj['retweeted_status']['user']['screen_name']\n",
    "\n",
    "            # Store the retweet text in 'retweeted_status-text'\n",
    "            tweet_obj['retweeted_status-text'] = \\\n",
    "                                        tweet_obj['retweeted_status']['text']\n",
    "    \n",
    "            if 'extended_tweet' in tweet_obj['retweeted_status']:\n",
    "                # Store the extended retweet text in \n",
    "                #'retweeted_status-extended_tweet-full_text'\n",
    "                tweet_obj['retweeted_status-extended_tweet-full_text'] = \\\n",
    "                tweet_obj['retweeted_status']['extended_tweet']['full_text']\n",
    "                \n",
    "        if 'quoted_status' in tweet_obj:\n",
    "            # Store the retweet user screen name in \n",
    "            #'retweeted_status-user-screen_name'\n",
    "            tweet_obj['quoted_status-user-screen_name'] = \\\n",
    "                            tweet_obj['quoted_status']['user']['screen_name']\n",
    "\n",
    "            # Store the retweet text in 'retweeted_status-text'\n",
    "            tweet_obj['quoted_status-text'] = \\\n",
    "                                            tweet_obj['quoted_status']['text']\n",
    "    \n",
    "            if 'extended_tweet' in tweet_obj['quoted_status']:\n",
    "                # Store the extended retweet text in \n",
    "                #'retweeted_status-extended_tweet-full_text'\n",
    "                tweet_obj['quoted_status-extended_tweet-full_text'] = \\\n",
    "                    tweet_obj['quoted_status']['extended_tweet']['full_text']\n",
    "        \n",
    "        ''' Place info'''\n",
    "        if 'place' in tweet_obj:\n",
    "            # Store the country code in 'place-country_code'\n",
    "            try:\n",
    "                tweet_obj['place-country'] = \\\n",
    "                                            tweet_obj['place']['country']\n",
    "                \n",
    "                tweet_obj['place-country_code'] = \\\n",
    "                                            tweet_obj['place']['country_code']\n",
    "                \n",
    "                tweet_obj['location-coordinates'] = \\\n",
    "                            tweet_obj['place']['bounding_box']['coordinates']\n",
    "            except: pass\n",
    "        \n",
    "        tweets_list.append(tweet_obj)\n",
    "        \n",
    "    return tweets_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {},
   "outputs": [],
   "source": [
    "def select_text(tweets):\n",
    "    ''' Assigns the main text to only one column depending\n",
    "        on whether the tweet is a RT/quote or not'''\n",
    "    \n",
    "    tweets_list = []\n",
    "    \n",
    "    # Iterate through each tweet\n",
    "    for tweet_obj in tweets:\n",
    "        \n",
    "        if 'retweeted_status-extended_tweet-full_text' in tweet_obj:\n",
    "            tweet_obj['text'] = \\\n",
    "                        tweet_obj['retweeted_status-extended_tweet-full_text']\n",
    "        \n",
    "        elif 'retweeted_status-text' in tweet_obj:\n",
    "            tweet_obj['text'] = tweet_obj['retweeted_status-text']\n",
    "            \n",
    "        elif 'extended_tweet-full_text' in tweet_obj:\n",
    "                    tweet_obj['text'] = tweet_obj['extended_tweet-full_text']\n",
    "                \n",
    "        tweets_list.append(tweet_obj)\n",
    "        \n",
    "    return tweets_list"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "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",
       "      <th>lang</th>\n",
       "      <th>user-location</th>\n",
       "      <th>place-country</th>\n",
       "      <th>place-country_code</th>\n",
       "      <th>location-coordinates</th>\n",
       "      <th>user-screen_name</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>@siironique @Raheem0971 @twm1s @lzylv \"il faut...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>zoltan16k</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>Nantes ce soir : moment historique, unique à n...</td>\n",
       "      <td>fr</td>\n",
       "      <td>FRANCE</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Cathylabb</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>Devant la dénégation du racisme endémique de l...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>samb_michele</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>Wallah ! Nique le ! Tue le ! Sale Gwer! ( sale...</td>\n",
       "      <td>fr</td>\n",
       "      <td>France - Paris</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>LeftyMsr</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>Devant la dénégation du racisme endémique de l...</td>\n",
       "      <td>fr</td>\n",
       "      <td>Paris, Ile-de-France</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bien_monsieur</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>Ou j’ai dis mme que le racisme est un truc de ...</td>\n",
       "      <td>fr</td>\n",
       "      <td>Rhône-Alpes, France</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>baabeegurl1</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>6</th>\n",
       "      <td>Wallah ! Nique le ! Tue le ! Sale Gwer! ( sale...</td>\n",
       "      <td>fr</td>\n",
       "      <td>Paris</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Carl_Marois</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>7</th>\n",
       "      <td>Devant la dénégation du racisme endémique de l...</td>\n",
       "      <td>fr</td>\n",
       "      <td>Troyes, France</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>Bruno__Chatenet</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>8</th>\n",
       "      <td>Devant la dénégation du racisme endémique de l...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>nagou_san</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>9</th>\n",
       "      <td>Wallah ! Nique le ! Tue le ! Sale Gwer! ( sale...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>NaN</td>\n",
       "      <td>tricolore78</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                                                text lang  \\\n",
       "0  @siironique @Raheem0971 @twm1s @lzylv \"il faut...   fr   \n",
       "1  Nantes ce soir : moment historique, unique à n...   fr   \n",
       "2  Devant la dénégation du racisme endémique de l...   fr   \n",
       "3  Wallah ! Nique le ! Tue le ! Sale Gwer! ( sale...   fr   \n",
       "4  Devant la dénégation du racisme endémique de l...   fr   \n",
       "5  Ou j’ai dis mme que le racisme est un truc de ...   fr   \n",
       "6  Wallah ! Nique le ! Tue le ! Sale Gwer! ( sale...   fr   \n",
       "7  Devant la dénégation du racisme endémique de l...   fr   \n",
       "8  Devant la dénégation du racisme endémique de l...   fr   \n",
       "9  Wallah ! Nique le ! Tue le ! Sale Gwer! ( sale...   fr   \n",
       "\n",
       "          user-location  place-country  place-country_code  \\\n",
       "0                  None            NaN                 NaN   \n",
       "1                FRANCE            NaN                 NaN   \n",
       "2                  None            NaN                 NaN   \n",
       "3        France - Paris            NaN                 NaN   \n",
       "4  Paris, Ile-de-France            NaN                 NaN   \n",
       "5   Rhône-Alpes, France            NaN                 NaN   \n",
       "6                 Paris            NaN                 NaN   \n",
       "7        Troyes, France            NaN                 NaN   \n",
       "8                  None            NaN                 NaN   \n",
       "9                  None            NaN                 NaN   \n",
       "\n",
       "   location-coordinates user-screen_name  \n",
       "0                   NaN        zoltan16k  \n",
       "1                   NaN        Cathylabb  \n",
       "2                   NaN     samb_michele  \n",
       "3                   NaN         LeftyMsr  \n",
       "4                   NaN    Bien_monsieur  \n",
       "5                   NaN      baabeegurl1  \n",
       "6                   NaN      Carl_Marois  \n",
       "7                   NaN  Bruno__Chatenet  \n",
       "8                   NaN        nagou_san  \n",
       "9                   NaN      tricolore78  "
      ]
     },
     "execution_count": 9,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# flatten tweets\n",
    "tweets = flatten_tweets(tweets_data)\n",
    "\n",
    "# select text\n",
    "tweets = select_text(tweets)\n",
    "columns = ['text', 'lang', 'user-location', 'place-country', \n",
    "           'place-country_code', 'location-coordinates', \n",
    "           'user-screen_name']\n",
    "\n",
    "# Create a DataFrame from `tweets`\n",
    "df_tweets = pd.DataFrame(tweets, columns=columns)\n",
    "# replaces NaNs by Nones\n",
    "#df_tweets.where(pd.notnull(df_tweets), None, inplace=True)\n",
    "df_tweets.head(10)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "<class 'pandas.core.frame.DataFrame'>\n",
      "RangeIndex: 100 entries, 0 to 99\n",
      "Data columns (total 7 columns):\n",
      "text                    100 non-null object\n",
      "lang                    100 non-null object\n",
      "user-location           54 non-null object\n",
      "place-country           0 non-null float64\n",
      "place-country_code      0 non-null float64\n",
      "location-coordinates    0 non-null float64\n",
      "user-screen_name        100 non-null object\n",
      "dtypes: float64(3), object(4)\n",
      "memory usage: 5.5+ KB\n"
     ]
    }
   ],
   "source": [
    "df_tweets.info()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# sample for quick check\n",
    "df_tweets_sample = df_tweets.copy()[:50]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <p><font color=lightsteelblue> &#128036; PART 3: Tweets analysis from MongoDB</font></p> <a id=\"dataanalysis\"></a>\n",
    "\n",
    "In this analysis, we decide to extract different informations from the tweets we collected:\n",
    "* popularity of the search interest (defined by you)\n",
    "* top 20 of the hashtags used in the collected tweets\n",
    "* Original content VS Retweet\n",
    "* Number of likes for the collected tweets\n",
    "* Type of content (text, photo, video)\n",
    "\n",
    "Depending of what you want to analyse, you can make these queries very. For instance if you look for information of an international event, you might want to extract and sort counrty data.\n",
    "<br><br><br>\n",
    "\n",
    "<p><font color=indianred>REQUIREMENT</font></p>\n",
    "\n",
    "For this step, you need to have the **mongod server running** (probably here: `MongoDB\\Server\\version\\bin\\mongod.exe`)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {},
   "outputs": [],
   "source": [
    "from collections import Counter\n",
    "import pymongo\n",
    "import re\n",
    "from pymongo import MongoClient\n",
    "\n",
    "# To import the collection to mongoDB:\n",
    "# mongoimport -d analysis -c brexit  --file brexit.json"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {},
   "outputs": [],
   "source": [
    "client = MongoClient('localhost', 27017)\n",
    "db = client['analysis']\n",
    "collection = db['loisecuriteglobale3']\n",
    "\n",
    "\n",
    "YOUR_SEARCH_INTEREST = \"Loi Securite Globale\"\n",
    "SEARCH_INTEREST = re.compile('stoploisecuriteglobale|loisecuriteglobale|pplsecuriteglobale')\n",
    "\n",
    "# output text file that contains the analysis\n",
    "outfile = open('tweets_analysis_output_file.txt', 'w')\n",
    "outfile.write('''\n",
    "    ###############\n",
    "    TWEETS ANALYSIS\n",
    "    ###############\n",
    "    ''')\n",
    "\n",
    "def find_popularity():\n",
    "    # ----------------------\n",
    "    # Mentions of __YOUR_SEARCH_INTEREST__ : x\n",
    "    # Total tweets:  xxxx\n",
    "    # -----------------------\n",
    "    outfile.write(\"\\nMENTIONS OF \" + YOUR_SEARCH_INTEREST + \":\\n--------------\")\n",
    "    occurence_of_search_interest = 0\n",
    "    # db_max = \n",
    "    # db_min = \n",
    "    total_valid = 0\n",
    "\n",
    "    cursor = collection.find({})\n",
    "    for doc in cursor:\n",
    "        # db_min += 1\n",
    "        # if db_min < db_max:\n",
    "        try:\n",
    "            text = doc['text']\n",
    "            total_valid += 1\n",
    "        except KeyError:\n",
    "            # print doc['_id']\n",
    "            pass\n",
    "        # print text\n",
    "        occurence_of_search_interest += len(SEARCH_INTEREST.findall(text.lower()))\n",
    "        # else:\n",
    "        #     break\n",
    "    outfile.write(\"\\nMentions of \" + YOUR_SEARCH_INTEREST + \" : \" + str(occurence_of_search_interest))\n",
    "    outfile.write(\"\\nTotal tweets: \" + str(total_valid))\n",
    "    outfile.write('\\n')\n",
    "    outfile.write('\\n')\n",
    "\n",
    "\n",
    "def top_20_hashtags():\n",
    "    # -------------------\n",
    "    # count hashtag\n",
    "    # -------------------\n",
    "    outfile.write(\"\\nTOP 20 HASHTAGS:\\n--------------\\n\")\n",
    "    '''\n",
    "    Traverse through all hashtags in every tweet\n",
    "    - hashtags are present in the attribute - entities.hashtags\n",
    "    - return value - array of hashtags, empty is no hashtag\n",
    "        {\"indices\":[x,y], \"text\": <hashtag-text-here>}\n",
    "\n",
    "    To generate word cloud: https://www.wordclouds.com/\n",
    "    '''\n",
    "    cursor = collection.find({})\n",
    "    # db_max = 10\n",
    "    # db_min = 0\n",
    "    hashtag_counter = Counter()\n",
    "    for doc in cursor:\n",
    "        # db_min += 1\n",
    "        # if db_min < db_max:\n",
    "        try:\n",
    "            hashtag_list = doc['entities']['hashtags']\n",
    "            if len(hashtag_list) > 0:\n",
    "                for ht in hashtag_list:\n",
    "                    hashtag_counter[ht['text'].lower()] += 1\n",
    "        except KeyError:\n",
    "            pass\n",
    "        # else:\n",
    "            # break\n",
    "    top_20 = hashtag_counter.most_common(20)\n",
    "\n",
    "    for word, count in top_20:\n",
    "        outfile.write(word + \": \" + str(count) + \"\\n\")\n",
    "    outfile.write(\"\\nTotal number of hashtags used: \" + str(sum(hashtag_counter.values())) + \"\\n\")\n",
    "    outfile.write('\\n')\n",
    "\n",
    "\n",
    "def original_vs_retweeted():\n",
    "\n",
    "    outfile.write(\"\\nTYPES OF TWEETS - ORIGINAL VS RETWEETED:\\n--------------\")\n",
    "    # Original tweets don't have the attribute\n",
    "    #   retweeted_status\n",
    "    # Original\n",
    "    # db.getCollection('loisecuriteglobale3').find({\"retweeted_status\":{$eq:null}},{}).length()\n",
    "    res = collection.count_documents({\"retweeted_status\":{'$eq':None}},{})\n",
    "    outfile.write(\"\\nNumber of original tweets: \" + str(res))\n",
    "    # Retweeted\n",
    "    # db.getCollection('loisecuriteglobale3').find({\"retweeted_status\":{$ne:null}},{}).length()\n",
    "    res = collection.count_documents({\"retweeted_status\":{'$ne':None}},{})\n",
    "    outfile.write(\"\\nNumber of retweeted tweets: \" + str(res))\n",
    "    outfile.write('\\n')\n",
    "    outfile.write('\\n')\n",
    "\n",
    "def fav_counts():\n",
    "    # find no of tweets greater than a number - ex: 50,000 => Outputs: 3\n",
    "    # raw mongo query\n",
    "    # db.getCollection('brexit').find({'retweeted_status.favorite_count':{$gt:30000}}).length()\n",
    "    outfile.write(\"\\nCOUNTS LIKES ON TWEETS:\\n--------------\")\n",
    "    times = [1, 100, 1000, 5000, 10000, 50000]\n",
    "    for value in times:\n",
    "        res = collection.count_documents({'retweeted_status.favorite_count':{'$gt':value}})\n",
    "        outfile.write(\"\\nNumber of tweets favorited more than \" + str(value) + \" times: \" + str(res))\n",
    "    outfile.write('\\n')\n",
    "    outfile.write('\\n')\n",
    "\n",
    "def tweet_type():\n",
    "\n",
    "    outfile.write(\"\\nTYPES OF TWEET CONTENT - TEXT / AUDIO / VIDEO\\n--------------\")\n",
    "    # only text: 4822\n",
    "    # db.getCollection('loisecuriteglobale3').find({'entities.media.type':{$eq:null}}).length()\n",
    "    res = collection.count_documents({'entities.media.type':{'$eq':None}})\n",
    "    outfile.write('\\nTweets with only text: ' + str(res))\n",
    "    # ----\n",
    "    # contains Photo: 178\n",
    "    # db.getCollection('loisecuriteglobale3').find({'extended_entities.media.type':'photo'}).length()\n",
    "    res = collection.count_documents({'extended_entities.media.type':'photo'})\n",
    "    outfile.write('\\nTweets with images: ' + str(res))\n",
    "    # ----\n",
    "    # contains Video: 19\n",
    "    # db.getCollection('loisecuriteglobale3').find({'extended_entities.media.type':'video'}).length()\n",
    "    res = collection.count_documents({'extended_entities.media.type':'video'})\n",
    "    outfile.write('\\nTweets with videos: ' + str(res))\n",
    "    # ----\n",
    "    outfile.write('')\n",
    "\n",
    "\n",
    "def main():\n",
    "    find_popularity()\n",
    "    top_20_hashtags()\n",
    "    original_vs_retweeted()\n",
    "    fav_counts()\n",
    "    tweet_type()\n",
    "\n",
    "if __name__ == '__main__':\n",
    "    main()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 35,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "\n",
      "    ###############\n",
      "    TWEETS ANALYSIS\n",
      "    ###############\n",
      "    \n",
      "MENTIONS OF Loi Securite Globale:\n",
      "--------------\n",
      "Mentions of Loi Securite Globale : 7\n",
      "Total tweets: 116\n",
      "\n",
      "\n",
      "TOP 20 HASHTAGS:\n",
      "--------------\n",
      "nantes: 4\n",
      "loisecuriteglobale: 4\n",
      "marchedeslibertés: 4\n",
      "stoploisecuriteglobale: 3\n",
      "france: 1\n",
      "macron: 1\n",
      "polqc: 1\n",
      "ensauvagement: 1\n",
      "\n",
      "Total number of hashtags used: 19\n",
      "\n",
      "\n",
      "TYPES OF TWEETS - ORIGINAL VS RETWEETED:\n",
      "--------------\n",
      "Number of original tweets: 31\n",
      "Number of retweeted tweets: 85\n",
      "\n",
      "\n",
      "COUNTS LIKES ON TWEETS:\n",
      "--------------\n",
      "Number of tweets favorited more than 1 times: 79\n",
      "Number of tweets favorited more than 100 times: 53\n",
      "Number of tweets favorited more than 1000 times: 11\n",
      "Number of tweets favorited more than 5000 times: 2\n",
      "Number of tweets favorited more than 10000 times: 0\n",
      "Number of tweets favorited more than 50000 times: 0\n",
      "\n",
      "\n",
      "TYPES OF TWEET CONTENT - TEXT / AUDIO / VIDEO\n",
      "--------------\n",
      "Tweets with only text: 110\n",
      "Tweets with images: 6\n",
      "Tweets with videos: 0\n"
     ]
    }
   ],
   "source": [
    "f = open('tweets_analysis_output_file.txt', 'r')\n",
    "file_contents = f.read()\n",
    "print (file_contents)\n",
    "f.close()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## <p><font color=lightsteelblue> &#129413; PART 4: Data visualisation </font></p>\n",
    "\n",
    "visualisation part based on [this code](https://github.com/ugis22/analysing_twitter/blob/master/Jupyter%20Notebook%20files/Interaction%20Network.ipynb)\n",
    "\n",
    "Reminder of what does a [tweet object](https://developer.twitter.com/en/docs/tweets/data-dictionary/overview/tweet-object) looks like !\n",
    "\n",
    "### a) Building the dataframe"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "Index(['_id', 'contributors', 'coordinates', 'created_at',\n",
       "       'display_text_range', 'entities', 'extended_entities', 'extended_tweet',\n",
       "       'favorite_count', 'favorited', 'filter_level', 'geo', 'id', 'id_str',\n",
       "       'in_reply_to_screen_name', 'in_reply_to_status_id',\n",
       "       'in_reply_to_status_id_str', 'in_reply_to_user_id',\n",
       "       'in_reply_to_user_id_str', 'is_quote_status', 'lang', 'place',\n",
       "       'possibly_sensitive', 'quote_count', 'quoted_status',\n",
       "       'quoted_status_id', 'quoted_status_id_str', 'quoted_status_permalink',\n",
       "       'reply_count', 'retweet_count', 'retweeted', 'retweeted_status',\n",
       "       'source', 'text', 'timestamp_ms', 'truncated', 'user'],\n",
       "      dtype='object')"
      ]
     },
     "execution_count": 15,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "import numpy as np\n",
    "import pandas as pd\n",
    "from scipy import stats\n",
    "from operator import itemgetter\n",
    "\n",
    "df_tweets = pd.read_json('loisecuriteglobale3.json', lines=True, dtype=False)\n",
    "df_tweets.columns"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Second dataframe to put information we want\n",
    "df_for_dataviz = pd.DataFrame(columns = [\"created_at\",\n",
    "                                         \"user_id\",\n",
    "                                         \"screen_name\",\n",
    "                                         \"followers_count\",\n",
    "                                         \"text\",\n",
    "                                         \"lang\",\n",
    "                                         \"in_reply_to_screen_name\",\n",
    "                                         \"in_reply_to_status_id\",\n",
    "                                         \"in_reply_to_user_id\",\n",
    "                                         \"retweeted_id\",\n",
    "                                         \"retweeted_screen_name\",\n",
    "                                         \"user_mentions_screen_name\",\n",
    "                                         \"user_mentions_id\",\n",
    "                                         \"id\",])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Columns that are going to be the same\n",
    "equal_columns = [\"created_at\", \"id\", \"text\", \"lang\"]\n",
    "df_for_dataviz[equal_columns] = df_tweets[equal_columns]"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Here we collect :\n",
    "* **User info**: Name, id, number of followers\n",
    "* **User mentions**: The `entities` object is a dictionary that contains, among other information, the `user_mentions` object that represents other Twitter users (`screen_name` and `id`) mentioned in the text of the Tweet.\n",
    "* **Retweets**: The retweets always contain two Tweet objects: The original Tweet being Retweeted is provided in a retweeted_status object. Inside, the user object for the account (screen_name and id) taking the Retweet action and the time of the Retweet.\n",
    "* **Replies**"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the basic information about user \n",
    "def get_basics(df_for_dataviz):\n",
    "    #df_for_dataviz[\"id\"] = df_tweets[\"user\"].apply(lambda x: x[\"id_str\"])\n",
    "    df_for_dataviz[\"screen_name\"] = df_tweets[\"user\"].apply(lambda x: x[\"screen_name\"])\n",
    "    df_for_dataviz[\"user_id\"] = df_tweets[\"user\"].apply(lambda x: x[\"id_str\"])\n",
    "    df_for_dataviz[\"followers_count\"] = df_tweets[\"user\"].apply(lambda x: x[\"followers_count\"][\"$numberInt\"])\n",
    "    return df_for_dataviz\n",
    "\n",
    "# Get the user mentions \n",
    "def get_usermentions(df_for_dataviz):\n",
    "    # Inside the tag 'entities' will find 'user mentions' and will get 'screen name' and 'id'\n",
    "    df_for_dataviz[\"user_mentions_screen_name\"] = df_tweets[\"entities\"].apply(lambda x: x[\"user_mentions\"][0][\"screen_name\"] if x[\"user_mentions\"] else np.nan)\n",
    "    df_for_dataviz[\"user_mentions_id\"] = df_tweets[\"entities\"].apply(lambda x: x[\"user_mentions\"][0][\"id_str\"] if x[\"user_mentions\"] else np.nan)\n",
    "    return df_for_dataviz\n",
    "\n",
    "# Get retweets\n",
    "def get_retweets(df_for_dataviz):\n",
    "    # Inside the tag 'retweeted_status' will find 'user' and will get 'screen name' and 'id'    \n",
    "    df_for_dataviz[\"retweeted_screen_name\"] = df_tweets[\"retweeted_status\"].apply(lambda x: x[\"user\"][\"screen_name\"] if x is not np.nan else np.nan)\n",
    "    df_for_dataviz[\"retweeted_id\"] = df_tweets[\"retweeted_status\"].apply(lambda x: x[\"user\"][\"id_str\"] if x is not np.nan else np.nan)\n",
    "    return df_for_dataviz\n",
    "\n",
    "# Get the information about replies\n",
    "def get_in_reply(df_for_dataviz):\n",
    "    # Just copy the 'in_reply' columns to the new dataframe\n",
    "    df_for_dataviz[\"in_reply_to_screen_name\"] = df_tweets[\"in_reply_to_screen_name\"]\n",
    "    df_for_dataviz[\"in_reply_to_status_id\"] = df_tweets[\"in_reply_to_status_id_str\"]\n",
    "    df_for_dataviz[\"in_reply_to_user_id\"]= df_tweets[\"in_reply_to_user_id_str\"]\n",
    "    return df_for_dataviz\n",
    "\n",
    "# Lastly fill the new dataframe with the important information\n",
    "def fill_df(df_for_dataviz):\n",
    "    get_basics(df_for_dataviz)\n",
    "    get_usermentions(df_for_dataviz)\n",
    "    get_retweets(df_for_dataviz)\n",
    "    get_in_reply(df_for_dataviz)\n",
    "    return df_for_dataviz"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Get the interactions between the different users\n",
    "def get_interactions(row):\n",
    "    # From every row of the original dataframe\n",
    "    # First we obtain the 'user_id' and 'screen_name'\n",
    "    user = row[\"user_id\"], row[\"screen_name\"]\n",
    "    # Be careful if there is no user id\n",
    "    if user[0] is None:\n",
    "        return (None, None), []\n",
    "    \n",
    "    # The interactions are going to be a set of tuples\n",
    "    interactions = set()\n",
    "    \n",
    "    # Add all interactions \n",
    "    # First, we add the interactions corresponding to replies adding the id and screen_name\n",
    "    interactions.add((row[\"in_reply_to_user_id\"], row[\"in_reply_to_screen_name\"]))\n",
    "    # After that, we add the interactions with retweets\n",
    "    interactions.add((row[\"retweeted_id\"], row[\"retweeted_screen_name\"]))\n",
    "    # And later, the interactions with user mentions\n",
    "    interactions.add((row[\"user_mentions_id\"], row[\"user_mentions_screen_name\"]))\n",
    "    \n",
    "    # Discard if user id is in interactions\n",
    "    interactions.discard((row[\"user_id\"], row[\"screen_name\"]))\n",
    "    # Discard all not existing values\n",
    "    interactions.discard((None, None))\n",
    "    # Return user and interactions\n",
    "    return user, interactions"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {
    "scrolled": true
   },
   "outputs": [
    {
     "data": {
      "text/html": [
       "<div>\n",
       "<style scoped>\n",
       "    .dataframe tbody tr th:only-of-type {\n",
       "        vertical-align: middle;\n",
       "    }\n",
       "\n",
       "    .dataframe tbody tr th {\n",
       "        vertical-align: top;\n",
       "    }\n",
       "\n",
       "    .dataframe thead th {\n",
       "        text-align: right;\n",
       "    }\n",
       "</style>\n",
       "<table border=\"1\" class=\"dataframe\">\n",
       "  <thead>\n",
       "    <tr style=\"text-align: right;\">\n",
       "      <th></th>\n",
       "      <th>created_at</th>\n",
       "      <th>user_id</th>\n",
       "      <th>screen_name</th>\n",
       "      <th>followers_count</th>\n",
       "      <th>text</th>\n",
       "      <th>lang</th>\n",
       "      <th>in_reply_to_screen_name</th>\n",
       "      <th>in_reply_to_status_id</th>\n",
       "      <th>in_reply_to_user_id</th>\n",
       "      <th>retweeted_id</th>\n",
       "      <th>retweeted_screen_name</th>\n",
       "      <th>user_mentions_screen_name</th>\n",
       "      <th>user_mentions_id</th>\n",
       "      <th>id</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>2020-12-15 18:05:49</td>\n",
       "      <td>1281669801249775617</td>\n",
       "      <td>zoltan16k</td>\n",
       "      <td>18</td>\n",
       "      <td>@siironique @Raheem0971 @twm1s @lzylv \"il faut...</td>\n",
       "      <td>fr</td>\n",
       "      <td>siironique</td>\n",
       "      <td>1338904281777319943</td>\n",
       "      <td>1210890516721631233</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>siironique</td>\n",
       "      <td>1210890516721631233</td>\n",
       "      <td>{'$numberLong': '1338908134245621761'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>2020-12-15 18:05:53</td>\n",
       "      <td>271439132</td>\n",
       "      <td>Cathylabb</td>\n",
       "      <td>1358</td>\n",
       "      <td>RT @Nantes_Revoltee: Nantes ce soir : moment h...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>726365373965152257</td>\n",
       "      <td>Nantes_Revoltee</td>\n",
       "      <td>Nantes_Revoltee</td>\n",
       "      <td>726365373965152257</td>\n",
       "      <td>{'$numberLong': '1338908150985072646'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>2020-12-15 18:05:55</td>\n",
       "      <td>717476046900166657</td>\n",
       "      <td>samb_michele</td>\n",
       "      <td>131</td>\n",
       "      <td>RT @sebastianjroche: Devant la dÃ©nÃ©gation du...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>2398454779</td>\n",
       "      <td>sebastianjroche</td>\n",
       "      <td>sebastianjroche</td>\n",
       "      <td>2398454779</td>\n",
       "      <td>{'$numberLong': '1338908158350258178'}</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>2020-12-15 18:05:55</td>\n",
       "      <td>1109875812</td>\n",
       "      <td>LeftyMsr</td>\n",
       "      <td>125</td>\n",
       "      <td>RT @GWGoldnadel: Wallah ! Nique le ! Tue le ! ...</td>\n",
       "      <td>fr</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>None</td>\n",
       "      <td>3373762791</td>\n",
       "      <td>GWGoldnadel</td>\n",
       "      <td>GWGoldnadel</td>\n",
       "      <td>3373762791</td>\n",
       "      <td>{'$numberLong': '1338908160602628096'}</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "           created_at              user_id   screen_name followers_count  \\\n",
       "0 2020-12-15 18:05:49  1281669801249775617     zoltan16k              18   \n",
       "1 2020-12-15 18:05:53            271439132     Cathylabb            1358   \n",
       "2 2020-12-15 18:05:55   717476046900166657  samb_michele             131   \n",
       "3 2020-12-15 18:05:55           1109875812      LeftyMsr             125   \n",
       "\n",
       "                                                text lang  \\\n",
       "0  @siironique @Raheem0971 @twm1s @lzylv \"il faut...   fr   \n",
       "1  RT @Nantes_Revoltee: Nantes ce soir : moment h...   fr   \n",
       "2  RT @sebastianjroche: Devant la dÃ©nÃ©gation du...   fr   \n",
       "3  RT @GWGoldnadel: Wallah ! Nique le ! Tue le ! ...   fr   \n",
       "\n",
       "  in_reply_to_screen_name in_reply_to_status_id  in_reply_to_user_id  \\\n",
       "0              siironique   1338904281777319943  1210890516721631233   \n",
       "1                    None                  None                 None   \n",
       "2                    None                  None                 None   \n",
       "3                    None                  None                 None   \n",
       "\n",
       "         retweeted_id retweeted_screen_name user_mentions_screen_name  \\\n",
       "0                None                  None                siironique   \n",
       "1  726365373965152257       Nantes_Revoltee           Nantes_Revoltee   \n",
       "2          2398454779       sebastianjroche           sebastianjroche   \n",
       "3          3373762791           GWGoldnadel               GWGoldnadel   \n",
       "\n",
       "      user_mentions_id                                      id  \n",
       "0  1210890516721631233  {'$numberLong': '1338908134245621761'}  \n",
       "1   726365373965152257  {'$numberLong': '1338908150985072646'}  \n",
       "2           2398454779  {'$numberLong': '1338908158350258178'}  \n",
       "3           3373762791  {'$numberLong': '1338908160602628096'}  "
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "df_for_dataviz = fill_df(df_for_dataviz)\n",
    "df_for_dataviz = df_for_dataviz.where((pd.notnull(df_for_dataviz)), None)\n",
    "df_for_dataviz.head(4)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### b) Knowing the graph\n",
    "\n",
    "we use [networkx](https://networkx.org/documentation/latest/index.html) now and we will use [pyvis](https://pyvis.readthedocs.io/en/latest/index.html) for interactive graphs (links to documentations)\n",
    "\n",
    "#### > Nodes and Edges\n",
    "<img src=\"graph_illus.png\" style=\"width: 200px;\"/>\n",
    "\n",
    "The most important functions of NetworkX are\n",
    "* `Graph()` to create the graph ([see doc here](https://networkx.org/documentation/stable/reference/classes/graph.html#networkx.Graph))\n",
    "* `add_node()` ([see doc here](https://networkx.org/documentation/stable/reference/classes/generated/networkx.Graph.add_node.html?highlight=add_node#networkx-graph-add-node))\n",
    "* `add_edge()` between nodes to connect ([see doc here](https://networkx.org/documentation/stable/reference/classes/generated/networkx.Graph.add_edge.html?highlight=add_edge#networkx.Graph.add_edge))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 105 nodes and 74 edges present in the Graph\n"
     ]
    }
   ],
   "source": [
    "# graphs visualisation\n",
    "import networkx as nx\n",
    "import matplotlib.pyplot as plt\n",
    "import seaborn as sns\n",
    "from pyvis.network import Network\n",
    "\n",
    "graph = nx.Graph()\n",
    "\n",
    "for index, tweet in df_for_dataviz.iterrows():\n",
    "    user, interactions = get_interactions(tweet)\n",
    "    user_id, user_name = user\n",
    "    tweet_id = tweet[\"id\"]\n",
    "    #tweet_sent = tweet[\"sentiment\"]\n",
    "    for interaction in interactions:\n",
    "        int_id, int_name = interaction\n",
    "        graph.add_edge(user_id, int_id, tweet_id=tweet_id)\n",
    "        \n",
    "\n",
    "        graph.nodes[user_id][\"name\"] = user_name\n",
    "        graph.nodes[int_id][\"name\"] = int_name\n",
    "\n",
    "nb_of_graph_nodes = graph.number_of_nodes()\n",
    "print(f\"There are {nb_of_graph_nodes} nodes and {graph.number_of_edges()} edges present in the Graph\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### > Degrees\n",
    "\n",
    "The **degree** of a node is the **number of edges that occur to that node**.<br>\n",
    "The maximum/minimum degrees of a graph is the max/min degrees of its nodes.<br>\n",
    "High degre = lots of connections / interconnections<br>\n",
    "Low degree = not really connected that much\n",
    "\n",
    "<img src=\"graph_illus2.png\" style=\"width: 500px;\" align=\"left\"/>\n",
    "here:<br>\n",
    "max degree = 9<br>\n",
    "min degree = 0<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The maximum degree of the Graph is 17\n",
      "The minimum degree of the Graph is 1\n",
      "The average degree of the nodes in the Graph is 1.4\n",
      "The most frequent degree of the nodes found in the Graph is 1\n"
     ]
    }
   ],
   "source": [
    "degrees = [val for (node, val) in graph.degree()]\n",
    "print(f\"The maximum degree of the Graph is {np.max(degrees)}\")   \n",
    "print(f\"The minimum degree of the Graph is {np.min(degrees)}\")\n",
    "print(f\"The average degree of the nodes in the Graph is {np.mean(degrees):.1f}\")  \n",
    "print(f\"The most frequent degree of the nodes found in the Graph is {stats.mode(degrees)[0][0]}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### > Connections\n",
    "A graph is connected if, for every pair of nodes, there is a path between them. (Like if you would draw the graph, you could keep your pen on the paper to draw it entirely with no jumps). To do so, most of the nodes should have at least a degree of two (except the \"ends\").<br>\n",
    "Let's use `nx.is_connected(graph)`"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The graph is not connected\n",
      "There are 31 connected components in the Graph\n"
     ]
    }
   ],
   "source": [
    "if nx.is_connected(graph):\n",
    "    print(\"The graph is connected\")\n",
    "else:\n",
    "    print(\"The graph is not connected\")\n",
    "\n",
    "print(f\"There are {nx.number_connected_components(graph)} connected components in the Graph\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### > Subgraphs\n",
    "\n",
    "We saw above the maximum degree of the graph. We will now explore this."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "There are 18 nodes and 17 edges present in the largest component of the Graph\n",
      "The largest subgraph is connected\n"
     ]
    }
   ],
   "source": [
    "def connected_component_subgraphs(G): # connected_component_subgraphs is depricated\n",
    "    for c in nx.connected_components(G):\n",
    "        yield G.subgraph(c)\n",
    "\n",
    "largest_subgraph = max(connected_component_subgraphs(graph), key=len)\n",
    "nb_of_subgraph_nodes = largest_subgraph.number_of_nodes()\n",
    "print(f\"There are {nb_of_subgraph_nodes} nodes and {largest_subgraph.number_of_edges()} \\\n",
    "edges present in the largest component of the Graph\")\n",
    "\n",
    "if nx.is_connected(largest_subgraph):\n",
    "    print(\"The largest subgraph is connected\")\n",
    "else:\n",
    "    print(\"The largest subgraph is not connected\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### > Clustering coefficient and Transitivity\n",
    "\n",
    "The **clustering coefficient**, a measure of the number of triangles in a graph : **\"Do my neighbors knows each other ?\"**\n",
    "<img src=\"graph_illus3.png\" style=\"width: 500px;\"/>\n",
    "\n",
    "The **transitivity** measures the density of loops of length three (triangles) in a network. ([more about transitivity](https://www.sci.unich.it/~francesc/teaching/network/transitivity.html))\n",
    "\n",
    "These two parameters are very important when analyzing social networks because it gives us an insight into how users tend to create tightly knot groups characterized by relatively high-dense ties."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The average clustering coefficient is 0.0 in the largest subgraph\n",
      "The transitivity of the largest subgraph is 0\n"
     ]
    }
   ],
   "source": [
    "print(f\"The average clustering coefficient is {nx.average_clustering(largest_subgraph)} in the largest subgraph\")\n",
    "print(f\"The transitivity of the largest subgraph is {nx.transitivity(largest_subgraph)}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### > Diameter and distance\n",
    "\n",
    "It's how far away one node is from another random node. Diameter represents the maximum distance between any pair of nodes while the average distance tells us the average distance between any two nodes in the network."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The diameter of our Graph is 2\n",
      "The average distance between any two nodes is 1.89\n"
     ]
    }
   ],
   "source": [
    "print(f\"The diameter of our Graph is {nx.diameter(largest_subgraph)}\")\n",
    "print(f\"The average distance between any two nodes is {nx.average_shortest_path_length(largest_subgraph):.2f}\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "#### > Network centrality\n",
    "\n",
    "**degree**: node with many connections<br>\n",
    "**closeness**: nodes are close to other nodes<br>\n",
    "**betweenness** high degree nodes connect other nodes<br>"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The node with id 2398454779 has a degree centrality of 1.00 which is the maximum of the Graph\n",
      "The node with id 2398454779 has a closeness centrality of 1.00 which is the maximum of the Graph\n",
      "The node with id 2398454779 has a betweenness centrality of 1.00 which is the maximum of the Graph\n"
     ]
    }
   ],
   "source": [
    "graph_centrality = nx.degree_centrality(largest_subgraph)\n",
    "max_de = max(graph_centrality.items(), key=itemgetter(1))\n",
    "graph_closeness = nx.closeness_centrality(largest_subgraph)\n",
    "max_clo = max(graph_closeness.items(), key=itemgetter(1))\n",
    "graph_betweenness = nx.betweenness_centrality(largest_subgraph, normalized=True, endpoints=False)\n",
    "max_bet = max(graph_betweenness.items(), key=itemgetter(1))\n",
    "\n",
    "degree_node = max_de[0]\n",
    "closeness_node = max_clo[0]\n",
    "betweenness_node = max_bet[0]\n",
    "print(f\"The node with id {degree_node} has a degree centrality of {max_de[1]:.2f} which is the maximum of the Graph\")\n",
    "print(f\"The node with id {closeness_node} has a closeness centrality of {max_clo[1]:.2f} which is the maximum of the Graph\")\n",
    "print(f\"The node with id {betweenness_node} has a betweenness centrality of {max_bet[1]:.2f} which is the maximum of the Graph\")"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### c1) Drawing the subgraph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Drawing of the subgraph\n",
    "node_and_degree = largest_subgraph.degree()\n",
    "colors_central_nodes = ['orange', 'blue', 'yellow']\n",
    "central_nodes = [degree_node, closeness_node, betweenness_node]\n",
    "\n",
    "pos = nx.spring_layout(largest_subgraph, k=0.05)\n",
    "\n",
    "plt.figure(figsize = (20,20))\n",
    "nx.draw(largest_subgraph, pos=pos, node_color=range(nb_of_subgraph_nodes), cmap=plt.cm.PiYG, edge_color=\"black\", linewidths=0.3, node_size=60, alpha=0.6, with_labels=False)\n",
    "nx.draw_networkx_nodes(largest_subgraph, pos=pos, nodelist=central_nodes, node_size=300, node_color=colors_central_nodes)\n",
    "plt.savefig('networkx_subgraph.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### c2) Drawing the graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "The node with id 2398454779 has a degree centrality of 0.16 which is the maximum of the Graph\n",
      "The node with id 2398454779 has a closeness centrality of 0.16 which is the maximum of the Graph\n",
      "The node with id 2398454779 has a betweenness centrality of 0.03 which is the maximum of the Graph\n"
     ]
    }
   ],
   "source": [
    "graph_centrality = nx.degree_centrality(graph)\n",
    "max_de = max(graph_centrality.items(), key=itemgetter(1))\n",
    "graph_closeness = nx.closeness_centrality(graph)\n",
    "max_clo = max(graph_closeness.items(), key=itemgetter(1))\n",
    "graph_betweenness = nx.betweenness_centrality(graph, normalized=True, endpoints=False)\n",
    "max_bet = max(graph_betweenness.items(), key=itemgetter(1))\n",
    "\n",
    "degree_node = max_de[0]\n",
    "closeness_node = max_clo[0]\n",
    "betweenness_node = max_bet[0]\n",
    "print(f\"The node with id {degree_node} has a degree centrality of {max_de[1]:.2f} which is the maximum of the Graph\")\n",
    "print(f\"The node with id {closeness_node} has a closeness centrality of {max_clo[1]:.2f} which is the maximum of the Graph\")\n",
    "print(f\"The node with id {betweenness_node} has a betweenness centrality of {max_bet[1]:.2f} which is the maximum of the Graph\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 1440x1440 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "node_and_degree = graph.degree()\n",
    "colors_central_nodes = ['orange', 'blue', 'yellow']\n",
    "central_nodes = [degree_node, closeness_node, betweenness_node]\n",
    "\n",
    "pos = nx.spring_layout(graph, k=0.05)\n",
    "\n",
    "\n",
    "plt.figure(figsize = (20,20))\n",
    "nx.draw(graph, pos=pos, node_color=range(nb_of_graph_nodes), cmap=plt.cm.PiYG, edge_color=\"black\", linewidths=0.3, node_size=60, alpha=0.6, with_labels=False)\n",
    "nx.draw_networkx_nodes(graph, pos=pos, nodelist=central_nodes, node_size=300, node_color=colors_central_nodes)\n",
    "plt.savefig('networkx_graph.png')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### d) Touching the graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "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>Source</th>\n",
       "      <th>Target</th>\n",
       "      <th>Weight</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>1281669801249775617</td>\n",
       "      <td>1210890516721631233</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>1281669801249775617</td>\n",
       "      <td>1268654112738103307</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>271439132</td>\n",
       "      <td>726365373965152257</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>726365373965152257</td>\n",
       "      <td>946377666</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>726365373965152257</td>\n",
       "      <td>1023945297838452737</td>\n",
       "      <td>0</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "                Source               Target  Weight\n",
       "0  1281669801249775617  1210890516721631233       0\n",
       "1  1281669801249775617  1268654112738103307       0\n",
       "2            271439132   726365373965152257       0\n",
       "3   726365373965152257            946377666       0\n",
       "4   726365373965152257  1023945297838452737       0"
      ]
     },
     "execution_count": 31,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# get Source and Targets\n",
    "source_lst = []\n",
    "target_lst = []\n",
    "weight_lst = []\n",
    "for e in graph.edges():\n",
    "    source,target = e\n",
    "    #print(source,target)\n",
    "    source_lst.append(source)\n",
    "    target_lst.append(target)\n",
    "    weight_lst.append(0)\n",
    "\n",
    "# DataFrame\n",
    "data_source_target = {'Source':source_lst,\n",
    "                      'Target':target_lst,\n",
    "                      'Weight':weight_lst}   \n",
    "df_graph = pd.DataFrame(data_source_target)\n",
    "df_graph.astype({'Source': 'str'}).dtypes\n",
    "df_graph.astype({'Target': 'str'}).dtypes\n",
    "df_graph.astype({'Weight': 'str'}).dtypes\n",
    "df_graph.to_csv('df_graph.csv', index=False)\n",
    "df_graph.head()"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/html": [
       "\n",
       "        <iframe\n",
       "            width=\"100%\"\n",
       "            height=\"450px\"\n",
       "            src=\"graph_dynamic.html\"\n",
       "            frameborder=\"0\"\n",
       "            allowfullscreen\n",
       "        ></iframe>\n",
       "        "
      ],
      "text/plain": [
       "<IPython.lib.display.IFrame at 0x188fc7874a8>"
      ]
     },
     "execution_count": 32,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "graph_dynamic_data = pd.read_csv(\"df_graph.csv\")\n",
    "\n",
    "graph_dynamic = nx.Graph()\n",
    "\n",
    "sources = graph_dynamic_data['Source']\n",
    "targets = graph_dynamic_data['Target']\n",
    "weights = graph_dynamic_data['Weight']\n",
    "edge_data = zip(sources, targets, weights)\n",
    "for e in edge_data:\n",
    "    src = e[0]\n",
    "    dst = e[1]\n",
    "    w = e[2]\n",
    "    graph_dynamic.add_node(src)\n",
    "    graph_dynamic.add_node(dst)\n",
    "    graph_dynamic.add_edge(src, dst, value=w)\n",
    "\n",
    "# add neighbor data to node hover data\n",
    "#for node in got_net.nodes:\n",
    "#    node[\"title\"] += \" Neighbors:<br>\" + \"<br>\".join(neighbor_map[node[\"id\"]])\n",
    "#    node[\"value\"] = len(neighbor_map[node[\"id\"]])\n",
    "\n",
    "# need to be improved because Network() should be defined first so I could give add_node() parameters\n",
    "# like here https://pyvis.readthedocs.io/en/latest/tutorial.html#adding-list-of-nodes-with-properties\n",
    "graph_dynamic_net = Network(height=\"450px\", width=\"100%\", bgcolor=\"#222222\", font_color=\"white\", notebook=True)\n",
    "graph_dynamic_net.barnes_hut() # set the physics layout of the network\n",
    "# populates the nodes and edges data structures\n",
    "graph_dynamic_net.from_nx(graph_dynamic)\n",
    "graph_dynamic_net.show(\"graph_dynamic.html\")"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.7.1"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 2
}