Task1-Graph_Analytics.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "id": "collect-election",
   "metadata": {},
   "source": [
    "# Task 1: Graph Analytics"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "apart-forth",
   "metadata": {},
   "source": [
    "## Extracting mention Graph"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "civil-typing",
   "metadata": {},
   "source": [
    "Importing all necessary libraries"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "black-botswana",
   "metadata": {},
   "outputs": [],
   "source": [
    "import csv\n",
    "import re\n",
    "import collections\n",
    "import operator\n",
    "import numpy as np\n",
    "import networkx as nx\n",
    "import matplotlib.pyplot as plt"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "macro-polls",
   "metadata": {},
   "source": [
    "A tweet object which contains the data extracted from the Twitter data"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "id": "published-balloon",
   "metadata": {},
   "outputs": [],
   "source": [
    "class tweet:\n",
    "  def __init__(self, t):\n",
    "    self.timestamp = t[0]\n",
    "    self.author = t[1]\n",
    "    self.content = t[2]"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "sweet-marble",
   "metadata": {},
   "source": [
    "Creating the dataset of tweets from the 'twitter.tsv' file. The .tsv file was read using the csv library, which also accepts tab seperated values."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "id": "orange-death",
   "metadata": {},
   "outputs": [],
   "source": [
    "with open(\"twitter.tsv\", encoding=\"utf8\") as fd:\n",
    "    \n",
    "    tweets = []\n",
    "    rd = csv.reader(fd, delimiter=\"\\t\", quotechar='\"')\n",
    "\n",
    "    for row in rd:\n",
    "        tweets.append(tweet(row))"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "nearby-enlargement",
   "metadata": {},
   "source": [
    "Creating a dictionary of edges. Using a dictionary gives us the ability to set a key pair and a corresponding value. This is useful because we can use the author of the tweet and the user mentioned as the key with the number of times mentioned as the value. This is important because the number of mentions is used as the weights on the edges of the graph.\n",
    "\n",
    "By using regex, we can extract just the user's alias without the extra @ symbol or any other characters. If a new (author, mentioned)key is found, it is added to the dictionary. If an existing (author, mentioned)key is found, the value is incremented instead. By iterating through only the tweets which contain an '@', it reduces running time and power."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "id": "entire-trigger",
   "metadata": {},
   "outputs": [],
   "source": [
    "edges = {}\n",
    "\n",
    "for t in tweets:\n",
    "    if \"@\" in t.content:\n",
    "        mentioned = re.findall(\"@(\\w+)\",t.content)\n",
    "        for m in mentioned:\n",
    "            key = tuple([t.author, m])\n",
    "            if key in edges.keys():\n",
    "                edges[key] += 1\n",
    "            else:\n",
    "                edges[key] = 1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "descending-edwards",
   "metadata": {},
   "source": [
    "Creating a Directed Graph using the edges found above. It is important to have a directed graph since user1 mentioning user2 does not mean that user2 mentioned user1."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "accessible-filing",
   "metadata": {},
   "outputs": [],
   "source": [
    "G = nx.DiGraph()\n",
    "\n",
    "for k, v in edges:\n",
    "    G.add_edge(k, v, weight=edges[(k,v)])\n",
    "    \n",
    "# Draw Full Graph\n",
    "#weights = [G[u][v]['weight'] for u,v in edges]\n",
    "#nx.draw(G, with_labels=True, node_size=60,font_size=10, width=weights)\n",
    "#plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "indonesian-drill",
   "metadata": {},
   "source": [
    "Creates a subgraph of the largest weakly connected component."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "id": "impressive-bulgaria",
   "metadata": {},
   "outputs": [],
   "source": [
    "largest_cc = max(nx.weakly_connected_components(G), key=len)\n",
    "SG = G.subgraph(largest_cc)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "comparative-burns",
   "metadata": {},
   "source": [
    "## Statistics for the graph"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "residential-ridge",
   "metadata": {},
   "source": [
    "#### Number Of Nodes\n",
    "Getting the number of nodes in the graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "id": "super-majority",
   "metadata": {},
   "outputs": [],
   "source": [
    "numberOfNodes = G.number_of_nodes()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "material-tokyo",
   "metadata": {},
   "source": [
    "#### Number Of Edges\n",
    "Getting the number of edges in the graph"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "id": "available-brother",
   "metadata": {},
   "outputs": [],
   "source": [
    "numberOfEdges = G.number_of_edges()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "executed-soundtrack",
   "metadata": {},
   "source": [
    "#### In-Degree\n",
    "Getting the number of vertices coming in to the node, from other connected nodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "id": "desperate-attempt",
   "metadata": {},
   "outputs": [],
   "source": [
    "inDegree = G.in_degree()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "hourly-understanding",
   "metadata": {},
   "source": [
    "#### Out-Degree\n",
    "Getting the number of vertices going out of the node, to other connected nodes"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "id": "stylish-million",
   "metadata": {},
   "outputs": [],
   "source": [
    "outDegree = G.out_degree()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "generous-norwegian",
   "metadata": {},
   "source": [
    "#### Degree Distribution\n",
    "The degree distribution is the probability distribution of the in and out degrees over the whole network"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "associate-synthesis",
   "metadata": {},
   "source": [
    "##### Histogram\n",
    "By plotting the graph using a log scale, we would be able to distinguish the results better since the graph wouldnt be so squashed."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "id": "approved-british",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(\"Degree Histogram\")\n",
    "plt.xlabel(\"Count\")\n",
    "plt.ylabel(\"Degree\")\n",
    "\n",
    "degrees = [G.degree(n) for n in G.nodes()]\n",
    "\n",
    "plt.hist(degrees,log='true')\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "turned-bathroom",
   "metadata": {},
   "source": [
    "##### Scatter Plot\n",
    "A scatter plot can give us a better indication as to how exactly the degree distribution is spread"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "id": "velvet-baptist",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {
      "needs_background": "light"
     },
     "output_type": "display_data"
    }
   ],
   "source": [
    "plt.title(\"Degree Scatter\")\n",
    "plt.xlabel(\"Rank\")\n",
    "plt.ylabel(\"Degree\")\n",
    "\n",
    "#creates a distribution of degrees\n",
    "degree=nx.degree_histogram(G)\n",
    "degree.sort(reverse=True)\n",
    "\n",
    "#Generate normalised values for y according to degree\n",
    "y=[z/float(sum(degree))for z in degree]\n",
    "\n",
    "#Generate X axis sequence, from 1 to maximum degree\n",
    "x=range(len(degree))\n",
    "plt.scatter(x,y,s=1,color=(1,0,0))\n",
    "\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "informational-typing",
   "metadata": {},
   "source": [
    "#### Average Path Length\n",
    "Average path length is the sum of path lengths between all pairs of nodes normalized by n*(n-1) where n is the number of nodes in G.\n",
    "Since this has to be worked out on a connected graph, the averagePathLength has to be calculated on the largest connected subgraph."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "id": "technical-procurement",
   "metadata": {},
   "outputs": [],
   "source": [
    "averagePathLength = nx.average_shortest_path_length(SG)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "brief-advocate",
   "metadata": {},
   "source": [
    "#### Global Clustering Coeefficient\n",
    "The Global Clustering Coefficient is the number of closed triplets (triangles) over the total number of triplets (both open and closed). A triplet consists of three nodes connected with 2 edges c. A triangle includes three closed overlapping triplets, one centered on each of the nodes. "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "id": "equivalent-commerce",
   "metadata": {},
   "outputs": [],
   "source": [
    "#The average clustering coefficient for the graph G\n",
    "clusteringCoeefficient = nx.average_clustering(G)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "id": "flexible-survivor",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Compute graph transitivity, the fraction of all possible triangles present in G.\n",
    "transitivity = nx.transitivity(G)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "executed-creek",
   "metadata": {},
   "source": [
    "### Printing Statistics"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "id": "romantic-vehicle",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number Of Nodes:  65760 \n",
      "\n",
      "Number Of Edges:  68839 \n",
      "\n",
      "Average Shortest Path Length (for largest connected graph):  0.013634224566790034 \n",
      "\n",
      "Clustering Coefficient:  0.020226636811185247 \n",
      "\n",
      "Transitivity:  0.01666282425331309 \n",
      "\n"
     ]
    }
   ],
   "source": [
    "print(\"Number Of Nodes: \", numberOfNodes, \"\\n\")\n",
    "print(\"Number Of Edges: \", numberOfEdges, \"\\n\")\n",
    "print(\"Average Shortest Path Length (for largest connected graph): \", averagePathLength, \"\\n\")\n",
    "print(\"Clustering Coefficient: \", clusteringCoeefficient, \"\\n\")\n",
    "print(\"Transitivity: \", transitivity, \"\\n\")\n",
    "#uncomment to show (commented since the gives a large output)\n",
    "#print(\"In Degree: \", inDegree, \"\\n\")\n",
    "#print(\"Out Degree: \", outDegree, \"\\n\")"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "mediterranean-cuisine",
   "metadata": {},
   "source": [
    "From the results above we can conclude that the graph is not totally connected, however there is a giant component which conatains the majority of the connected nodes. The rest form smaller components. There are also some user with a very high in degree but have an out degree of nearly zero."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "intense-bachelor",
   "metadata": {},
   "source": [
    "## Determining the Top 10 users"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "following-crowd",
   "metadata": {},
   "source": [
    "Prints the keys and corresponding values from a dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "id": "subject-quilt",
   "metadata": {},
   "outputs": [],
   "source": [
    "def printDict(d):\n",
    "    for key in d:\n",
    "        print(key, '\\t:\\t', d[key])"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "presidential-syndrome",
   "metadata": {},
   "source": [
    "Prints the top 'n' number of users from a dictionary"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "id": "allied-hunger",
   "metadata": {},
   "outputs": [],
   "source": [
    "def printTopN(d, n):\n",
    "    counter = 1\n",
    "    for key in d:\n",
    "        print(key, '\\t:\\t', d[key])\n",
    "        if (counter >= n):\n",
    "            break\n",
    "        counter +=1"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "binding-lithuania",
   "metadata": {},
   "source": [
    "Sorts a dictionary according to the value in descending order"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "id": "running-citizenship",
   "metadata": {},
   "outputs": [],
   "source": [
    "def sortDictDesc(d):\n",
    "    sorted_d = dict( sorted(d.items(), key=operator.itemgetter(1),reverse=True))\n",
    "    return sorted_d"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "fatty-spanking",
   "metadata": {},
   "source": [
    "The degree centrality for a node v is the fraction of nodes it is connected to."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "id": "directed-headset",
   "metadata": {},
   "outputs": [],
   "source": [
    "degreeCentrality = nx.degree_centrality(G)\n",
    "degreeCentrality = sortDictDesc(degreeCentrality)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "august-publication",
   "metadata": {},
   "source": [
    "Closeness centrality of a node u is the reciprocal of the sum of the shortest path distances from u to all n-1 other nodes."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "id": "accurate-trader",
   "metadata": {},
   "outputs": [],
   "source": [
    "closenessCentrality = nx.closeness_centrality(G)\n",
    "closenessCentrality = sortDictDesc(closenessCentrality)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "abandoned-wallpaper",
   "metadata": {},
   "source": [
    "Betweenness centrality of a node v is the sum of the fraction of all-pairs shortest paths that pass through v"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "id": "sensitive-tablet",
   "metadata": {},
   "outputs": [],
   "source": [
    "betweennessCentrality = nx.betweenness_centrality(G, normalized=True)\n",
    "betweennessCentrality = sortDictDesc(betweennessCentrality)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "id": "medieval-operations",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Degree Centrality\n",
      "\n",
      "theiphoneblog \t:\t 0.037637433659270976\n",
      "ryanbarr \t:\t 0.0270989522346751\n",
      "mashable \t:\t 0.019191289405252512\n",
      "squarespace \t:\t 0.015572012956401406\n",
      "scottbourne \t:\t 0.015419942517374049\n",
      "ScanCafe \t:\t 0.015404735473471312\n",
      "iphone_dev \t:\t 0.013245335239282837\n",
      "tweetmeme \t:\t 0.011876701288036619\n",
      "TweetDeck \t:\t 0.009397953131890692\n",
      "kevinrose \t:\t 0.00783162760990891\n",
      "\n",
      "\n",
      "Closeness Centrality\n",
      "\n",
      "theiphoneblog \t:\t 0.027427802451198077\n",
      "ryanbarr \t:\t 0.026914331574782308\n",
      "mashable \t:\t 0.02663036318100742\n",
      "iphone_dev \t:\t 0.02255799832103937\n",
      "reneritchie \t:\t 0.0208254544501898\n",
      "MuscleNerd \t:\t 0.020822118432203253\n",
      "chrispirillo \t:\t 0.020199392874692457\n",
      "twittelator \t:\t 0.019479052477141254\n",
      "palmsolo \t:\t 0.01925193466205196\n",
      "backlon \t:\t 0.01921324516522264\n",
      "\n",
      "\n",
      "Betweenness Centrality\n",
      "\n",
      "theiphoneblog \t:\t 0.0004253726048982866\n",
      "filjedi \t:\t 0.0002926259536491623\n",
      "msproductions \t:\t 0.00029029017378310397\n",
      "lemonkey \t:\t 0.0002066224788538912\n",
      "dreadpiratepj \t:\t 0.00019356789904008073\n",
      "serban \t:\t 0.00019314523704854261\n",
      "iphonespaz \t:\t 0.00019188912230257968\n",
      "webaddict \t:\t 0.0001876675900566202\n",
      "johnbiggs \t:\t 0.00018355324575507683\n",
      "dgshaw \t:\t 0.00018180979394719454\n"
     ]
    }
   ],
   "source": [
    "print(\"Degree Centrality\\n\")\n",
    "printTopN(degreeCentrality, 10)\n",
    "print(\"\\n\\nCloseness Centrality\\n\")\n",
    "printTopN(closenessCentrality, 10)\n",
    "print(\"\\n\\nBetweenness Centrality\\n\")\n",
    "printTopN(betweennessCentrality, 10)"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "hollywood-webmaster",
   "metadata": {},
   "source": [
    "Directionality is important in this graph. This is because if user1 mentions user2, it doesnt mean that user2 mentions user1. The functions used to calulate degree, closeness and betweenness centrality all work for undirected, directed as well as multigraphs. Therefore there were no problems faced with regards to directionality."
   ]
  },
  {
   "cell_type": "markdown",
   "id": "primary-administration",
   "metadata": {},
   "source": [
    "## Visualise the Graph"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "coordinate-climate",
   "metadata": {},
   "source": [
    "Plots the whole graph with all the edges and labels. However this results in an undistinguishable graph."
   ]
  },
  {
   "cell_type": "raw",
   "id": "suburban-alpha",
   "metadata": {},
   "source": [
    "weights = [G[u][v]['weight'] for u,v in edges]\n",
    "\n",
    "node_sizes = []\n",
    "for n in closenessCentrality:\n",
    "    #print(closenessCentrality[n])\n",
    "    node_sizes.append(10000*closenessCentrality[n])\n",
    "\n",
    "nx.draw(G, node_size=node_sizes,with_labels=True, font_size=4, width=weights)\n",
    "plt.show()"
   ]
  },
  {
   "cell_type": "markdown",
   "id": "final-canvas",
   "metadata": {},
   "source": [
    "Instead of plotting the whole graph, we can plot the subgraph of strongly connected components of the whole graph. This will only show us the most important components of the graph. The size of the nodes also depends on their betweeness centrality, where the larger the centrality, the larger the size of the node will be. The width of the vertices/arrows also depends on the weight of the edge, meaning the number of mentions. By using the kamada_kawai drawing method we can give the graph a better visualised structure."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "id": "specialized-sauce",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 432x288 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "largest_cc = max(nx.strongly_connected_components(G), key=len)\n",
    "SG = G.subgraph(largest_cc)\n",
    "\n",
    "weights = [G[u][v]['weight'] for u,v in edges]\n",
    "\n",
    "node_sizes = []\n",
    "for n in SG.nodes():\n",
    "    node_sizes.append(1000000*betweennessCentrality[n])\n",
    "\n",
    "nx.draw_kamada_kawai(SG, node_size=node_sizes, with_labels=True, font_size=8, width=weights)\n",
    "plt.show()"
   ]
  }
 ],
 "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": "3.8.2"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}