P7_link_prediction_PPGIA.ipynb

{
 "cells": [
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "sQj7DPTViFDu"
   },
   "source": [
    "<p> <img src='images/Unifor_vertBlue.png' align=\"right\" width=\"200px\"> </p>\n",
    "\n",
    "# Atividade 7 - Predição de Links no Grafo de Coautoria do PPGIA<br>\n",
    "AIRES, Marcos. CARUBBI, Ricardo. SIQUEIRA, Fernando. Alunos de Análise de Dados em Grafos.<br>\n",
    "Professor orientador: Prof. Dr. Raimir Holanda.<br>\n",
    "Programa de Pós-Graduação em Informática Aplicada. Universidade de Fortaleza - UNIFOR.\n",
    "\n",
    "\n"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "Lt3MUMGF7s6G"
   },
   "source": [
    "## Resumo"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "WNdpY6ti6vST"
   },
   "source": [
    "Diversas informações sobre a produção cientı́fica nacional podem ser encontradas na Plataforma Lattes e, assim como as redes sociais, pode-se aplicar os mesmos conceitos para predição de links na rede social acadêmica. Esses links são identificados como coautorias, ou relações entre os pesquisadores na produção de uma publicação, e sua predição facilita e otimiza a produção científica. Este projeto comparou diversas métricas baseadas em similiaridade e proximidade em uma rede de coautoria dos professores do Programa de Pós-graduação em Informática Aplicada (PPGIA) da Universidade de Fortaleza (Unifor). Verificou-se que esta rede de coautoria, gerada em 1989 na abertura da referida pós-graduação, intensificou a colaboração acadêmica entre professores das respectivas áreas de concentração. Contudo apenas um link foi gerado no período de 2014 a 2021 e abordado como um problema de predição de links. Os resultados das diversas métricas não foram promissores sendo apenas um índice favorável à predição da referida coautoria. Conclui-se que para o problema de predição de coautorias em redes sociais acadêmicas é necessário analisar outras métricas e estratégias além das baseadas na estrutura da rede."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "U5AvIlJ7TPJu"
   },
   "source": [
    "## Configurações"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 1,
   "metadata": {},
   "outputs": [],
   "source": [
    "def set_node_community(G, communities):\n",
    "    '''Add community to node attributes'''\n",
    "    for c, v_c in enumerate(communities):\n",
    "        for v in v_c:\n",
    "            # Add 1 to save 0 for external edges\n",
    "            G.nodes[v]['community'] = c + 1\n",
    "            \n",
    "def set_edge_community(G):\n",
    "    '''Find internal edges and add their community to their attributes'''\n",
    "    for v, w, in G.edges:\n",
    "        if G.nodes[v]['community'] == G.nodes[w]['community']:\n",
    "            # Internal edge, mark with community\n",
    "            G.edges[v, w]['community'] = G.nodes[v]['community']\n",
    "        else:\n",
    "            # External edge, mark as 0\n",
    "            G.edges[v, w]['community'] = 0\n",
    "            \n",
    "def get_color(i, r_off=1, g_off=1, b_off=1):\n",
    "    r0, g0, b0 = 0, 0, 0\n",
    "    n = 16\n",
    "    low, high = 0.1, 0.9\n",
    "    span = high - low\n",
    "    r = low + span * (((i + r_off) * 3) % n) / (n - 1)\n",
    "    g = low + span * (((i + g_off) * 5) % n) / (n - 1)\n",
    "    b = low + span * (((i + b_off) * 7) % n) / (n - 1)\n",
    "    return (r, g, b)\n",
    "\n",
    "# ignoring warnings\n",
    "import warnings\n",
    "warnings.simplefilter(\"ignore\")\n",
    "\n",
    "import matplotlib.cm as cm\n",
    "import community as community_louvain\n",
    "def graph_louvain (G,ano):\n",
    "    \n",
    "    # Best partition\n",
    "    partition = community_louvain.best_partition(G, weight='weight', random_state=1503)\n",
    "    n_partitions = max(partition.values())+1\n",
    "    print(f'Número de partições: {n_partitions}')\n",
    "    \n",
    "    C_lvn=[]\n",
    "    for n in range(n_partitions):\n",
    "        C_lvn.append([k for k, v in partition.items() if v==n])\n",
    "    print(f'Nós por partição: {C_lvn}')\n",
    "    \n",
    "    # Set node and edge communities\n",
    "    set_node_community(G, C_lvn)\n",
    "    set_edge_community(G)\n",
    "\n",
    "    # Set community color for nodes\n",
    "    node_color = [get_color(G.nodes[v]['community']) for v in G.nodes]\n",
    "\n",
    "    # Set community edges\n",
    "    C_edges = []\n",
    "    for n in range(n_partitions+1):\n",
    "        C_edges.append([(v, w) for v, w in G.edges if G.edges[v, w]['community'] == n])\n",
    "\n",
    "    # Set community colors\n",
    "    C_colors = []\n",
    "    for c in C_edges:\n",
    "        C_colors.append([get_color(G.edges[e]['community']) for e in c])\n",
    "\n",
    "    # Set community weights\n",
    "    C_weights = []\n",
    "    for l in C_edges:\n",
    "        C_weights.append([G.get_edge_data(n1,n2)['weight'] for n1,n2 in l])   \n",
    "        \n",
    "    cmap = cm.get_cmap('spring', n_partitions)\n",
    "    \n",
    "    # Figure\n",
    "    plt.figure(figsize=(10,8))\n",
    "    plt.title(f'Detecção de Comunidades em {ano} por Louvain')\n",
    "\n",
    "    nx.draw_networkx(G,\n",
    "                     pos=pos,\n",
    "                     node_size=0,\n",
    "                     font_size=10,\n",
    "                     edgelist=C_edges[0],\n",
    "                     width=C_weights[0],\n",
    "                     edge_color='silver')\n",
    "\n",
    "    # Draw C1 edges\n",
    "    nx.draw_networkx(G,\n",
    "                     pos=pos,\n",
    "                     node_size=0,\n",
    "                     font_size=10,\n",
    "                     edgelist=C_edges[1],\n",
    "                     width=C_weights[1],\n",
    "                     edge_color=C_colors[1])\n",
    "\n",
    "    # Draw C2 edges\n",
    "    nx.draw_networkx(G,\n",
    "                     pos=pos,\n",
    "                     node_size=0,\n",
    "                     font_size=10,\n",
    "                     edgelist=C_edges[2],\n",
    "                     width=C_weights[2],\n",
    "                     edge_color=C_colors[2])\n",
    "\n",
    "    # Draw C3 edges\n",
    "    nx.draw_networkx(G,\n",
    "                     pos=pos,\n",
    "                     node_size=0,\n",
    "                     font_size=10,\n",
    "                     edgelist=C_edges[3],\n",
    "                     width=C_weights[3],\n",
    "                     edge_color=C_colors[3])\n",
    "\n",
    "    # Draw edge labels\n",
    "    labels = nx.get_edge_attributes(G,'weight')\n",
    "    nx.draw_networkx_edge_labels(G,\n",
    "                                 pos,\n",
    "                                 edge_labels=labels,\n",
    "                                 font_size=8, \n",
    "                                 verticalalignment='center_baseline', \n",
    "                                 alpha=.5)\n",
    "\n",
    "    nx.draw_networkx_labels(G, \n",
    "                            pos, \n",
    "                            alpha=.75, \n",
    "                            font_size=10)\n",
    "\n",
    "    nx.draw_networkx_nodes(G, \n",
    "                           pos, \n",
    "                           partition.keys(), \n",
    "                           node_size=300,\n",
    "                           node_color=node_color)\n",
    "\n",
    "    plt.show()\n",
    "    \n",
    "from matplotlib import pyplot as plt\n",
    "def graph_mono(G,ano):\n",
    "    \n",
    "    # Create figure\n",
    "    plt.figure(figsize=(8,8))\n",
    "    plt.title(f'Grafo Monopartido Ponderado PPGIA - Período de 1989 até {ano}')\n",
    "    # Get weights\n",
    "    weight = [G.edges[e]['weight'] for e in G.edges]\n",
    "    # Draw edges, nodes, and labels\n",
    "    nx.draw_networkx(G,\n",
    "                     node_color='silver',\n",
    "                     edge_color='silver',\n",
    "                     font_size=10,\n",
    "                     width=weight,\n",
    "                     pos=pos)\n",
    "    # Draw labels\n",
    "    labels = nx.get_edge_attributes(G,'weight')\n",
    "    nx.draw_networkx_edge_labels(G,\n",
    "                                 pos,\n",
    "                                 edge_labels=labels,\n",
    "                                 font_size=8, \n",
    "                                 verticalalignment='center_baseline', \n",
    "                                 alpha=.5)\n",
    "    nx.draw_networkx_labels(G, \n",
    "                            pos, \n",
    "                            font_size=10, \n",
    "                            alpha=.75)\n",
    "    plt.show()\n",
    "    \n",
    "def partition(G):\n",
    "    C_lvn=[]\n",
    "    partition = community_louvain.best_partition(H, random_state=1503)\n",
    "    n_partitions = max(partition.values())+1\n",
    "    for n in range(n_partitions):\n",
    "        C_lvn.append([k for k, v in partition.items() if v==n])\n",
    "    return C_lvn\n",
    "\n",
    "def avalCN(G,n1,n2):\n",
    "    cn = sorted(nx.common_neighbors(G,n1,n2))\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'Nó(s) vizinho(s): {cn}')\n",
    "    print(f'Número de nó(s) vizinho(s) em comum: {len(cn)}\\n')\n",
    "\n",
    "import math\n",
    "def salton_index(G,n1,n2):\n",
    "    kx = G.degree[n1]\n",
    "    ky = G.degree[n2]\n",
    "    si = len(list(nx.common_neighbors(G,n1,n2)))/math.sqrt(kx*ky)\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'salton index: {si:.4f}\\n')\n",
    "\n",
    "def sorensen_index(G,n1,n2):\n",
    "    kx = G.degree[n1]\n",
    "    ky = G.degree[n2]\n",
    "    si = 2*len(list(nx.common_neighbors(G,n1,n2)))/(kx+ky)\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'sorensen index: {si:.4f}\\n')\n",
    "\n",
    "def hub_promoted_index(G,n1,n2):\n",
    "    kx = G.degree[n1]\n",
    "    ky = G.degree[n2]\n",
    "    hub = len(list(nx.common_neighbors(G,n1,n2)))/min(kx,ky)\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'hub promoted index: {hub:.4f}\\n')\n",
    "    \n",
    "def hub_depressed_index(G,n1,n2):\n",
    "    kx = G.degree[n1]\n",
    "    ky = G.degree[n2]\n",
    "    hub = len(list(nx.common_neighbors(G,n1,n2)))/max(kx,ky)\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'hub depressed index: {hub:.4f}\\n')\n",
    "    \n",
    "def leicht_holme_newman_index(G,n1,n2):\n",
    "    kx = G.degree[n1]\n",
    "    ky = G.degree[n2]\n",
    "    lhn = len(list(nx.common_neighbors(G,n1,n2)))/(kx*ky)\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'leicht holme newman index: {lhn:.4f}\\n')\n",
    "    \n",
    "def preferential_attachment(G,n1,n2):\n",
    "    kx = G.degree[n1]\n",
    "    ky = G.degree[n2]\n",
    "    pa = kx*ky\n",
    "    print(f'Nós avaliados:{n1,n2}')\n",
    "    print(f'preferential attachment index: {pa:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "2am6jY5sBEaZ"
   },
   "source": [
    "## Carregamento do Grafo"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {
    "executionInfo": {
     "elapsed": 7265,
     "status": "ok",
     "timestamp": 1621474098154,
     "user": {
      "displayName": "Ricardo Carubbi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjFIu0VqofZXcKxSonHFG93xYoAWQRXL7z9LIkWBA=s64",
      "userId": "03831168603992390324"
     },
     "user_tz": 180
    },
    "id": "DYsMYIc8AtcS"
   },
   "outputs": [],
   "source": [
    "# Criação do grafo não direcionado vazio\n",
    "import datetime as dt\n",
    "import networkx as nx\n",
    "A = nx.Graph() # Grafo até o ano de 2010\n",
    "B = nx.Graph() # Grafo até o ano de 2014\n",
    "C = nx.Graph() # Grafo do período completo\n",
    "\n",
    "# Professores do PPGIA\n",
    "PPGIA = ['Placido Rogerio Pinheiro','Adriano Bessa Albuquerque','Americo Tadeu Falcone Sampaio',\n",
    "         'Joao Batista Furlan Duarte','Joao Jose Vasco Peixoto Furtado','Maria Andreia Formico Rodrigues',\n",
    "         'Maria Elizabeth Sucupira Furtado','Nabor das Chagas Mendonça','Napoleao Vieira Nepomuceno',\n",
    "         'Raimir Holanda Filho','Tarcisio Haroldo Cavalcante Pequeno','Vladia Celia Monteiro Pinheiro']\n",
    "\n",
    "# Carregamento dos dados no grafo\n",
    "with open('dataset/ppgia_coautorias.csv') as f:\n",
    "    next(f)\n",
    "    for row in f:\n",
    "        # Definição dos nós\n",
    "        linha = row.strip().split(\";\")\n",
    "        ano = dt.datetime.strptime(linha[0],'%Y')\n",
    "        artigo = linha[1]\n",
    "        autor = linha[2]\n",
    "        ano_p1 = dt.datetime(2010,1,1)\n",
    "        ano_p2 = dt.datetime(2014,1,1)\n",
    "        if autor in PPGIA:\n",
    "            # Grafo de todo o período\n",
    "            C.add_edge(autor, artigo)\n",
    "            # Definição dos atributos dos nós\n",
    "            C.nodes[autor]['autor'] = 1\n",
    "            C.nodes[artigo]['autor'] = 0\n",
    "            \n",
    "            if ano<=ano_p1:\n",
    "                # Grafo até o ano de 2010\n",
    "                A.add_edge(autor, artigo)\n",
    "                A.nodes[autor]['autor'] = 1\n",
    "                A.nodes[artigo]['autor'] = 0\n",
    "                \n",
    "            if ano<=ano_p2:\n",
    "                # Grafo até o ano de 2014\n",
    "                B.add_edge(autor, artigo)\n",
    "                B.nodes[autor]['autor'] = 1\n",
    "                B.nodes[artigo]['autor'] = 0 "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {
    "executionInfo": {
     "elapsed": 7257,
     "status": "ok",
     "timestamp": 1621474098168,
     "user": {
      "displayName": "Ricardo Carubbi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjFIu0VqofZXcKxSonHFG93xYoAWQRXL7z9LIkWBA=s64",
      "userId": "03831168603992390324"
     },
     "user_tz": 180
    },
    "id": "gecvUW3QG_HM"
   },
   "outputs": [],
   "source": [
    "# Renomea os labels dos nós como inteiros sequenciais e atribuir o nome completo como atributo\n",
    "grafo=[]\n",
    "for N in [A,B,C]:\n",
    "    grafo.append(nx.relabel.convert_node_labels_to_integers(N,\n",
    "                                                            first_label=0, \n",
    "                                                            label_attribute='nome'))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {
    "executionInfo": {
     "elapsed": 7240,
     "status": "ok",
     "timestamp": 1621474098175,
     "user": {
      "displayName": "Ricardo Carubbi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjFIu0VqofZXcKxSonHFG93xYoAWQRXL7z9LIkWBA=s64",
      "userId": "03831168603992390324"
     },
     "user_tz": 180
    },
    "id": "e8XOUS9X59iG"
   },
   "outputs": [],
   "source": [
    "# Gera e define o atributo 'bipartite' para identificação dos grupos de autores e artigos\n",
    "bp_autor=[]\n",
    "bp_artigos=[]\n",
    "for N in grafo:\n",
    "    bp_autor.append([v for v in N.nodes if N.nodes[v][\"autor\"] == 1])\n",
    "    bp_artigos.append([v for v in N.nodes if N.nodes[v][\"autor\"] == 0])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {
    "executionInfo": {
     "elapsed": 7240,
     "status": "ok",
     "timestamp": 1621474098175,
     "user": {
      "displayName": "Ricardo Carubbi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjFIu0VqofZXcKxSonHFG93xYoAWQRXL7z9LIkWBA=s64",
      "userId": "03831168603992390324"
     },
     "user_tz": 180
    },
    "id": "e8XOUS9X59iG"
   },
   "outputs": [],
   "source": [
    "# Gera e define o atributo 'bipartite' para identificação dos grupos de autores e artigos\n",
    "autores = []\n",
    "artigos = []\n",
    "for N in grafo:\n",
    "    autores.append([v for v in N.nodes if N.nodes[v][\"autor\"] == 1])\n",
    "    artigos.append([v for v in N.nodes if N.nodes[v][\"autor\"] == 0])  "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de autores A: 12\n",
      "Número de autores B: 12\n",
      "Número de autores C: 12\n"
     ]
    }
   ],
   "source": [
    "# Gera q quantidade de autores por grafo temporal:\n",
    "str_grafo=['A','B','C']\n",
    "for i,j in zip(str_grafo,autores):\n",
    "    print(f'Número de autores {i}: {len(j)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 7,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de artigos A: 114\n",
      "Número de artigos B: 175\n",
      "Número de artigos C: 304\n"
     ]
    }
   ],
   "source": [
    "# Gera q quantidade de artigos por grafo temporal:\n",
    "for i,j in zip(str_grafo,artigos):\n",
    "    print(f'Número de artigos {i}: {len(j)}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 8,
   "metadata": {
    "colab": {
     "base_uri": "https://localhost:8080/"
    },
    "executionInfo": {
     "elapsed": 11988,
     "status": "ok",
     "timestamp": 1621474103051,
     "user": {
      "displayName": "Ricardo Carubbi",
      "photoUrl": "https://lh3.googleusercontent.com/a-/AOh14GjFIu0VqofZXcKxSonHFG93xYoAWQRXL7z9LIkWBA=s64",
      "userId": "03831168603992390324"
     },
     "user_tz": 180
    },
    "id": "UTW9wro5IaYW",
    "outputId": "2307df01-fdb2-45c1-8f2c-2bdc38596b46",
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Identificação de professores ordenados pelo número de trabalhos publicados\n",
    "nome_grau=[]\n",
    "for N in grafo:\n",
    "    n_grau=nx.get_node_attributes(N, \"nome\")\n",
    "    nome_grau.append([(f'nó {x}', n_grau[x], N.degree[x]) for x in sorted(autores[0],\n",
    "                                                           key=lambda x:N.degree[x],\n",
    "                                                           reverse=True)])"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 9,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Grafo A:\n",
      "[('nó 0', 'Placido Rogerio Pinheiro', 46), ('nó 7', 'Maria Elizabeth Sucupira Furtado', 16), ('nó 11', 'Maria Andreia Formico Rodrigues', 14), ('nó 24', 'Joao Batista Furlan Duarte', 14), ('nó 9', 'Joao Jose Vasco Peixoto Furtado', 12), ('nó 44', 'Adriano Bessa Albuquerque', 9), ('nó 19', 'Nabor das Chagas Mendonça', 8), ('nó 22', 'Raimir Holanda Filho', 4), ('nó 39', 'Americo Tadeu Falcone Sampaio', 3), ('nó 69', 'Napoleao Vieira Nepomuceno', 3), ('nó 37', 'Vladia Celia Monteiro Pinheiro', 2), ('nó 119', 'Tarcisio Haroldo Cavalcante Pequeno', 2)]\n",
      "\n",
      "Grafo B:\n",
      "[('nó 0', 'Placido Rogerio Pinheiro', 73), ('nó 7', 'Maria Elizabeth Sucupira Furtado', 24), ('nó 11', 'Maria Andreia Formico Rodrigues', 21), ('nó 9', 'Joao Jose Vasco Peixoto Furtado', 18), ('nó 24', 'Joao Batista Furlan Duarte', 16), ('nó 44', 'Adriano Bessa Albuquerque', 15), ('nó 19', 'Nabor das Chagas Mendonça', 12), ('nó 22', 'Raimir Holanda Filho', 10), ('nó 69', 'Napoleao Vieira Nepomuceno', 7), ('nó 119', 'Tarcisio Haroldo Cavalcante Pequeno', 6), ('nó 37', 'Vladia Celia Monteiro Pinheiro', 4), ('nó 39', 'Americo Tadeu Falcone Sampaio', 4)]\n",
      "\n",
      "Grafo C:\n",
      "[('nó 0', 'Placido Rogerio Pinheiro', 123), ('nó 9', 'Joao Jose Vasco Peixoto Furtado', 42), ('nó 7', 'Maria Elizabeth Sucupira Furtado', 40), ('nó 11', 'Maria Andreia Formico Rodrigues', 36), ('nó 44', 'Adriano Bessa Albuquerque', 28), ('nó 24', 'Joao Batista Furlan Duarte', 25), ('nó 22', 'Raimir Holanda Filho', 23), ('nó 19', 'Nabor das Chagas Mendonça', 19), ('nó 37', 'Vladia Celia Monteiro Pinheiro', 11), ('nó 69', 'Napoleao Vieira Nepomuceno', 11), ('nó 119', 'Tarcisio Haroldo Cavalcante Pequeno', 9), ('nó 39', 'Americo Tadeu Falcone Sampaio', 6)]\n",
      "\n"
     ]
    }
   ],
   "source": [
    "# Gera um dicionário com nós, nomes dos autores e número de publicações (grau do nó). \n",
    "for i,j in zip(str_grafo,nome_grau):\n",
    "    print(f'Grafo {i}:\\n{j}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "RLHRetpB9jH3"
   },
   "source": [
    "## Grafo Monopartido Ponderado de Coautoria."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {
    "id": "t56PL1ObzFvL",
    "tags": []
   },
   "source": [
    "> Um grafo monopartido U (ou projeção de U) é obtido pela conexão entre dois nós U que compartilham o mesmo nó V na representação bipartida de U\". Pode-se definir duas projeções para um grafo bipartido, respectivamente uma projeção de U em V e de V em U.\n",
    "\n",
    "Barabási, Albert-László ; Pósfai, Márton: Network science. Cambridge : Cambridge University Press, 2016."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 10,
   "metadata": {
    "tags": []
   },
   "outputs": [],
   "source": [
    "# Calculate layout\n",
    "# pos = nx.spring_layout(G, weight='weight', k=0.1)\n",
    "from pickle import dump, load\n",
    "# dump(pos, open('output/pos.pkl', \"wb\"))\n",
    "pos=load(open('output/pos.pkl', \"rb\"))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 11,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Create graph weighted projected\n",
    "from networkx import bipartite\n",
    "mono=[]\n",
    "for i,j in zip(grafo,autores):\n",
    "    mono.append(bipartite.weighted_projected_graph(i,j))"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 12,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de Arestas: 13\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de Arestas: 18\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    },
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de Arestas: 19\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 576x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Gera os grafos monopartido ponderado de acordo com os anos e evidencia a variação de arestas.\n",
    "anos=[ano_p1.year,ano_p2.year,ano.year]\n",
    "for i,j in zip(mono,anos):\n",
    "    print(f'Número de Arestas: {len(i.edges)}')\n",
    "    graph_mono(i,j)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Detecção de Comunidades"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Algoritmo de Louvain"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "> The Louvain algorithm named after its inventors’ home town in Belgium, is a heuristic algorithm for approximately maximizing modularity over divisions of a network into any number of communities. The Louvain algorithm is an agglomerative algorithm, which works by taking single nodes and joining them into groups, then joining groups with other groups, and so forth, in an effort to find the configuration with highest modularity.\n",
    "\n",
    "Mark Newman. 2010. Networks: An Introduction. Oxford University Press, Inc., USA.<br>\n",
    "Blondel, V. D., Guillaume, J.-L., Lambiotte, R., and Lefebvre, E., Fast unfolding of communities in large networks, J. Stat. Mech. 2008, P10008 (2008)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 13,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Atribuição de variáveis aos grafos\n",
    "H=mono[1]\n",
    "I=mono[2]"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 14,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de partições: 3\n",
      "Nós por partição: [[0, 24, 44, 69], [7, 9, 22, 37, 119], [11, 19, 39]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "graph_louvain(H,2014)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 15,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Número de partições: 3\n",
      "Nós por partição: [[11, 19, 39], [7, 9, 22, 37, 119], [0, 24, 44, 69]]\n"
     ]
    },
    {
     "data": {
      "image/png": "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\n",
      "text/plain": [
       "<Figure size 720x576 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "graph_louvain(I,2021)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 16,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "[[{0: 2, 7: 1, 9: 1, 11: 3, 19: 3, 22: 1, 24: 2, 37: 1, 39: 0, 44: 2, 69: 2, 119: 1}], [{0: 0, 7: 2, 9: 2, 11: 1, 19: 1, 22: 2, 24: 0, 37: 2, 39: 1, 44: 0, 69: 0, 119: 2}], [{0: 0, 7: 2, 9: 2, 11: 1, 19: 1, 22: 2, 24: 0, 37: 2, 39: 1, 44: 0, 69: 0, 119: 2}]]\n"
     ]
    }
   ],
   "source": [
    "import community\n",
    "dendo=[]\n",
    "for i in mono:\n",
    "    dendo.append(community.generate_dendrogram(i))\n",
    "print(dendo)"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "## Link Prediction"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Existem inúmeras métricas para a predição de links, dentre elas estão as métricas baseadas em similaridade (ou proximidade). As técnicas que utilizam essas métricas calculam para cada par de vértices da rede, uma pontuação (score), que representa sua similaridade. Os altos valores de pontuação indicam alta probabilidade de existência do link."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "T. Zhou, L. Lu, Y.-C. Zhang. Predicting missing links via local information. Eur. Phys. J. B 71 (2009) 623. https://arxiv.org/pdf/0901.0553.pdf"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 17,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Aresta a ser predita: [(0, 22)]\n",
      "Nós da aresta a ser predita: [0, 22]\n"
     ]
    }
   ],
   "source": [
    "# Aresta a ser predita pela diferença entre as arestas dos grafos de 2021 e 2014.\n",
    "aresta_pred = [e for e in I.edges if e not in H.edges]\n",
    "print(f'Aresta a ser predita: {aresta_pred}')\n",
    "nos_pred = [e for l in aresta_pred for e in l]\n",
    "print(f'Nós da aresta a ser predita: {nos_pred}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 18,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós da comunidade 1 em 2014: [0, 24, 44, 69]\n",
      "Nós da comunidade 2 em 2014: [7, 9, 22, 37, 119]\n",
      "Nós da comunidade 3 em 2014: [11, 19, 39]\n"
     ]
    }
   ],
   "source": [
    "# Nós das comunidades identificados por Louvain.\n",
    "C1,C2,C3 = partition(H)\n",
    "print(f'Nós da comunidade 1 em 2014: {C1}')\n",
    "print(f'Nós da comunidade 2 em 2014: {C2}')\n",
    "print(f'Nós da comunidade 3 em 2014: {C3}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 19,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós possíveis: [37, 119]\n"
     ]
    }
   ],
   "source": [
    "# Identificar os nós da comunidade C2 que poderiam formar aresta com nó 22.\n",
    "nos_comp=[]\n",
    "for i in C2:\n",
    "    if i not in list(H.neighbors(C2[2]))+[nos_pred[1]]:\n",
    "        nos_comp.append(i)\n",
    "print(f'Nós possíveis: {nos_comp}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 20,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[0, 37, 119]"
      ]
     },
     "execution_count": 20,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "# Nós a serem avaliados\n",
    "nos=[nos_pred[0]]+nos_comp\n",
    "nos"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice de Vizinhos em Comum"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Vizinhos em Comum (Common Neighbors) define a quantidade de vizinhos em comum entre um par de nós. À medida que a quantidade de vizinhos em comum cresce, a chance dos dois nós terem um link entre si aumenta.\n",
    "\n",
    "$$s_{xy} = |Γ (x) ∩ Γ (y)|$$\n",
    "\n",
    "Sendo $Γ(x)$ o conjunto de vizinhos do vértice x e $|Γ(x)|$ a quantidade de vizinhos do\n",
    "vértice x, isto é, o grau do vértice."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 21,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "Nó(s) vizinho(s): [7]\n",
      "Número de nó(s) vizinho(s) em comum: 1\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "Nó(s) vizinho(s): [7, 9]\n",
      "Número de nó(s) vizinho(s) em comum: 2\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "Nó(s) vizinho(s): [9]\n",
      "Número de nó(s) vizinho(s) em comum: 1\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    avalCN(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, através do número de vizinhos comuns, a probabilidade da geração da aresta (22,37) é duas vezes maior que a aresta (22,0) ou da aresta (22,119)."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Coeficiente de Jaccard"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Coeficiente de Jaccard [1] é uma métrica de similaridade que normaliza a quantidade de vizinhos em comum.\n",
    "\n",
    "$$s_{xy}=\\frac{|Γ (x) ∩ Γ (y)|}{|Γ (x) ∪ Γ (y)|}$$\n",
    "\n",
    "1. P. Jaccard, Bulletin de la Societe Vaudoise des Science Naturelles 37, 547 (1901)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 22,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "jaccard index: 0.1667\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "jaccard index: 0.5000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "jaccard index: 0.2500\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.jaccard_coefficient(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'jaccard index: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de Jaccard, a probabilidade da geração da aresta (22,37) é 299,94% maior que a aresta (22,0), assim como a aresta (22,119) com 149,97%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice de Adamic-Adar"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Índice de Adamic-Adar [1] é uma métrica de similaridade que pondera a vizinhança em comum entre os vértices. Para isso, atribui maior peso aos vizinhos em comum com menor quantidade de arestas.\n",
    "\n",
    "$$S_{xy}=\\sum_{w \\in \\Gamma(u) \\cap \\Gamma(v)} \\frac{1}{\\log |\\Gamma(w)|}$$\n",
    "\n",
    "onde $Γ(u)$ denota um conjunto de vizinhos de $u$. Este índice leva à divisão zero para nós conectados apenas por meio de loops. Ele deve ser usado quando não houver self-loops presentes.\n",
    "\n",
    "1. L. A. Adamic, E. Adar, Social Networks 25, 211 (2003)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 23,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "Adamic-Adar Index: 0.6213\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "Adamic-Adar Index: 1.1794\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "Adamic-Adar Index: 0.5581\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.adamic_adar_index(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'Adamic-Adar Index: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de Adamic-Adar, a probabilidade da geração da aresta (22,37) é 189.83% em relação à aresta (22,0), assim como a aresta (22,119) com 89.83%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Conexão Preferencial"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Conexão Preferencial (Preferential Attachment) [1][2] é uma métrica onde as chances de um vértice estar conectado a outro vértice em uma rede é proporcional ao produto do número de vizinhos que cada um possui. A ideia é que novos relacionamentos têm mais chances de correr com pessoas que têm muitos relacionamentos (conhecido como o rico fica mais rico) [3].\n",
    "\n",
    "$$s_{xy} = {{k_{(x)} \\times k_{(y)}}}$$\n",
    "\n",
    "1. ADAMIC, L.; ADAR, E. Friends and neighbors on the web. Social Networks, v. 25, p.211–230, 2001.\n",
    "2. BARABáSI, A. L. et al. Evolution of the social network of scientific collaborations. Physica A: Statistical Mechanics and its Applications, v. 311, n. 3-4, p. 590–614, 2002.\n",
    "3. HASAN, M.; ZAKI, M. A survey of link prediction in social networks. In: AGGARWAL, C. C. (Ed.). Social Network Data Analytics. [S.l.]: Springer US, 2011."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 24,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "preferential attachment index: 10.0000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "preferential attachment index: 8.0000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "preferential attachment index: 6.0000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    preferential_attachment(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice Preferential Attachment, a probabilidade da geração da aresta (22,37) é de 80% em relação à aresta (22,0), assim como a aresta (22,119) de 60%. Este índice de predição atesta com a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice de Salton"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Salton ou Similaridade Cosseno [1] é uma medida da similaridade entre dois vetores que mede o cosseno do ângulo entre os mesmos.\n",
    "\n",
    "$$s_{xy}=\\frac{|Γ_(x) ∩ Γ (y)|}{\\sqrt{k_{(x)} \\times k_{(y)}}}$$\n",
    "\n",
    "onde $k(x) = |Γ (x)|$ denota o grau de $x$.\n",
    "\n",
    "1. G. Salton, M. J. McGill, Introduction to Modern Information Retrieval (MuGraw-Hill, Auckland, 1983)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 25,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "salton index: 0.3162\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "salton index: 0.7071\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "salton index: 0.4082\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    salton_index(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de Salton, a probabilidade da geração da aresta (22,37) é 223,63% maior que a aresta (22,0), assim como a aresta (22,119) com 129,10%."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice de Sørensen"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A métrica de S∅rensen [1] é usada para medir a similaridade entre duas amostras, frequentemente empregado em análises de comunidades ecológicas.\n",
    "\n",
    "$$s_{xy}=\\frac{|Γ_(x) ∩ Γ (y)|}{{k_{(x)} + k_{(y)}}}$$\n",
    "\n",
    "1. T. Sørensen, Biol. Skr. 5, 1 (1948)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 26,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "sorensen index: 0.2857\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "sorensen index: 0.6667\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "sorensen index: 0.4000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    sorensen_index(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de Sørensen, a probabilidade da geração da aresta (22,37) é 233,36% em relação à aresta (22,0), assim como a aresta (22,119) com 140,01%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice Hub Promoted"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "A métrica Hub Promoted Index [1] foi proposta para medir a sobreposição topológica de pares de substratos de redes metabólicas.\n",
    "\n",
    "$$s_{xy}=\\frac{|Γ_(x) ∩ Γ (y)|}{\\min({k_{(x)},k_{(y))}}}$$\n",
    "\n",
    "1. E. Ravasz, A. L. Somera, D. A. Mongru, Z. N. Oltvai, A.-L. Barabási, Science 297, 1553 (2002)."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "hub promoted index: 0.5000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "hub promoted index: 1.0000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "hub promoted index: 0.5000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    hub_promoted_index(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice Hub Promoted, a probabilidade da geração da aresta (22,37) é 250,00% em relação à aresta (22,0), assim como a aresta (22,119) com 166,65%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice Hub Depressed"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Hub Depressed Index [1] é a métrica oposta a métrica anterior.\n",
    "\n",
    "$$s_{xy}=\\frac{|Γ_(x) ∩ Γ (y)|}{\\max({k_{(x)},k_{(y))}}}$$\n",
    "\n",
    "1. ZHOU, T.; L, L.; ZHANG, Y.-C. Predicting missing links via local information. The European Physical Journal B, Springer-Verlag, v. 71, n. 4, p. 623–630, 2009."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 28,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "hub depressed index: 0.2000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "hub depressed index: 0.5000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "hub depressed index: 0.3333\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    hub_depressed_index(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice Hub Promoted, a probabilidade da geração da aresta (22,37) é 250,00% em relação à aresta (22,0), assim como a aresta (22,119) com 166,65%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice de Leicht-Holme-Newman"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O índice de Leicht-Holme-Newman [1] é uma medida de similaridade proposta com base no conceito de que dois vértices são semelhantes se os seus vizinhos na rede também são semelhantes.\n",
    "\n",
    "$$s_{xy}=\\frac{|Γ_(x) ∩ Γ (y)|}{{k_{(x)} \\times k_{(y)}}}$$\n",
    "\n",
    "onde o denominador, $k(x)×k(y)$, é proporcional ao número esperado de viznhos comuns dos nós $x$ e $y$.\n",
    "\n",
    "1. LEICHT, E. A.; HOLME, P.; NEWMAN, M. E. J. Vertex similarity in networks. Phys. Rev. E, American Physical Society, v. 73, p. 026120, Feb 2006."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "leicht holme newman index: 0.1000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "leicht holme newman index: 0.2500\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "leicht holme newman index: 0.1667\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    leicht_holme_newman_index(H,nos_pred[1],nos[i])"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de Leicht-Holme-Newman, a probabilidade da geração da aresta (22,37) é 250% em relação à aresta (22,0), assim como a aresta (22,119) com 166,67%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Alocação de Recurso"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Resource Allocation é uma métrica que, dado um par de vértices não conectados diretamente, tenta mensurar a transmissão de recursos entre estes vértices. Para um vértice $x$ enviar um recurso para o vértice $y$, tem-se a utilização dos vértices vizinhos para transmissão. No caso mais simples, assume-se que cada transmissor possui uma unidade de recurso e a distribuição é feita igualmente para todos os vizinhos. Portanto a similaridade é dada pela quantidade de recursos que $y$ recebe de $x$.\n",
    "\n",
    "$$S_{xy}=\\sum_{w \\in \\Gamma(u) \\cap \\Gamma(v)} \\frac{1}{|\\Gamma(w)|}$$\n",
    "\n",
    "onde $Γ(u)$ denota o conjunto de nós vizinhos de $u$.\n",
    "\n",
    "1. OU, Q. et al. Power-law strength-degree correlation from resource-allocation dynamics on weighted networks. Physical Review E, v. 75, n. 2, p. 021102, 2007."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "Resource Allocation index: 0.2000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "Resource Allocation index: 0.3667\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "Resource Allocation index: 0.1667\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.resource_allocation_index(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'Resource Allocation index: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice Resource Allocation Index, a probabilidade da geração da aresta (22,37) é 183.35% em relação à aresta (22,0), assim como a aresta (22,119) com 83.35%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice Cn Soundarajan Hopcroft"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "O índice Cn Soundarajan Hopcroft computa o número de nós vizinhos comuns de $u$ e $v$ e bonifica a cada nó vizinho comum que pertence a mesma comunidade de $u$ e $v$.\n",
    "\n",
    "$$S_{xy}=|\\Gamma(u) \\cap \\Gamma(v)| + \\sum_{w \\in \\Gamma(u) \\cap \\Gamma(v)} f(w)$$\n",
    "\n",
    "onde $f(w)$ igual a 1 se $w$ pertence a mesma comunidade de $u$ e $v$ ou igual a zero se contraário e $Γ(u)$ denota o conjunto de nós vizinhos de $u$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "cn Soundarajan Hopcroft index: 1.0000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "cn Soundarajan Hopcroft index: 4.0000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "cn Soundarajan Hopcroft index: 2.0000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.cn_soundarajan_hopcroft(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'cn Soundarajan Hopcroft index: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de CN Soundarajan Hopcroft, a probabilidade da geração da aresta (22,37) é 400% em relação à aresta (22,0), assim como a aresta (22,119) com 200%. Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Índice Ra Soundarajan Hopcroft"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calcula o índice de alocação de recursos de todos os pares de nós usando informações da comunidade. Seja os nós $u$ e $v$, esta função calcula o índice de alocação de recursos considerando apenas vizinhos comuns pertencentes à mesma comunidade que $u$ e $v$\n",
    "\n",
    "$$S_{xy}=\\sum_{w \\in \\Gamma(u) \\cap \\Gamma(v)} \\frac{f(w)}{|\\Gamma(w)|}$$\n",
    "\n",
    "onde $f(w)$ igual a 1 se $w$ pertence a mesma comunidade de $u$ e $v$ ou igual a zero se contraário e $Γ(u)$ denota o conjunto de nós vizinhos de $u$."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "Ra Index Soundarajan Hopcroft: 0.0000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "Ra Index Soundarajan Hopcroft: 0.3667\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "Ra Index Soundarajan Hopcroft: 0.1667\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.ra_index_soundarajan_hopcroft(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'Ra Index Soundarajan Hopcroft: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como este índice prevê apenas links em nós de mesma comunidade, e portanto, não atesta a existência da aresta (22,0) em 2021, pois seus repsectivos nós pertences à comunidades diferentes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Intra e Inter Cluster"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calcula a proporção de vizinhos comuns dentro e entre clusters de todos os pares de nós. Seja os nós $u$ e $v$, se um nó vizinho comum $w$ pertencer à mesma comunidade de $u$ e $v$, $w$ é considerado como um nó vizinho comum dentro do cluster de $u$ e $v$. Caso contrário, é considerado nós vizinho comum intercluster de $u$ e $v$. A proporção entre o tamanho do conjunto de nós vizinhos comuns dentro e entre os clusters é definida como a métrica WIC."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 33,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "WIC index: 0.0000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "WIC index: 2000.0000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "WIC index: 1000.0000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.within_inter_cluster(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'WIC index: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Como este índice prevê apenas links em nós de mesma comunidade, e portanto, não atesta a existência da aresta (22,0) em 2021, pois seus respectivos nós pertences à comunidades diferentes."
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "### Centralidade de Vizinho Communs"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Calcula a pontuação do algoritmo Common Neighbor Centrality (CCPA) de todos os pares de nós. A pontuação CCPA de $u$ e $v$ é definida como:\n",
    "\n",
    "$$ \\ alpha \\ cdot (| \\ Gamma (u) {\\ cap} ^ {} \\ Gamma (v) |) + (1- \\ alpha) \\ cdot \\ frac {N} {{d} _ {uv}} $$\n",
    "\n",
    "onde $Γ(u)$ é o conjunto de nós vizinhos de $u$, $Γ(v)$ é o conjunto de nós vizinhos de $v$, $α$ é o parâmetro que varia entre [0,1], N corresponde ao número total de nós no grafo e $duv$ é a distância mais curta entre os nós $u$ e $v$.\n",
    "\n",
    "Este algoritmo é baseado em duas propriedades vitais dos nós: o número de nós vizinhos comuns e sua centralidade."
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "metadata": {
    "tags": []
   },
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Nós avaliados:(22, 0)\n",
      "CCPA index: 2.0000\n",
      "\n",
      "Nós avaliados:(22, 37)\n",
      "CCPA index: 2.8000\n",
      "\n",
      "Nós avaliados:(22, 119)\n",
      "CCPA index: 2.0000\n",
      "\n"
     ]
    }
   ],
   "source": [
    "for i in range(len(nos)):\n",
    "    print(f'Nós avaliados:{nos_pred[1],nos[i]}')\n",
    "    for u, v, p in nx.common_neighbor_centrality(H,[(nos_pred[1],nos[i])]):\n",
    "        print(f'CCPA index: {p:.4f}\\n')"
   ]
  },
  {
   "cell_type": "markdown",
   "metadata": {},
   "source": [
    "Pode-se concluir que, baseado no índice de Common Neighbor Centrality, a probabilidade da geração da aresta (22,37) é 140% em relação à aresta (22,0). A probabilidade da aresta (22,119) é igual a da aresta (22,0). Este índice de predição não atesta a existência da aresta (22,0) em 2021."
   ]
  }
 ],
 "metadata": {
  "colab": {
   "authorship_tag": "ABX9TyMfW1Pu/AX6LUODwKDf4ioE",
   "collapsed_sections": [
    "U5AvIlJ7TPJu"
   ],
   "mount_file_id": "1a-QZLAbo2zm1Ek4iAAoyvuUMMtQTMz-Q",
   "name": "P5b_PPGIA_Bipartite.ipynb",
   "provenance": [
    {
     "file_id": "1BXy-0M6BK6N2FqwWSTpKilMGbfDz-vKX",
     "timestamp": 1619978937071
    },
    {
     "file_id": "18K2KdMPy5W18A6MGNnboEL0_AwD6AAmU",
     "timestamp": 1619890692028
    },
    {
     "file_id": "1k52FgW77tPmDOj5iYR-x5QpiNLHhYe6z",
     "timestamp": 1619889556910
    }
   ],
   "toc_visible": true
  },
  "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": "3.8.8"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}