dummy-bipartite-network-with-evaluation.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 2,
   "metadata": {},
   "outputs": [],
   "source": [
    "import pandas as pd\n",
    "import networkx as nx\n",
    "import itertools\n",
    "import matplotlib.pyplot as plt\n",
    "import random\n",
    "import numpy as np\n",
    "from sklearn.model_selection import KFold\n",
    "from sklearn.metrics import roc_auc_score, precision_recall_curve, average_precision_score, auc\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 3,
   "metadata": {},
   "outputs": [],
   "source": [
    "# Define functions here\n",
    "\n",
    "def visualize_bipartite_graph(G, d_nodes, t_nodes):\n",
    "    pos = {}\n",
    "    pos.update((node, (1, index)) for index, node in enumerate(d_nodes))  # positions for d nodes\n",
    "    pos.update((node, (2, index)) for index, node in enumerate(t_nodes))  # positions for t nodes\n",
    "\n",
    "    # Draw nodes and edges\n",
    "    plt.figure(figsize=(12, 8))\n",
    "    nx.draw(G, pos, with_labels=True, node_color=['skyblue' if node in d_nodes else 'lightgreen' for node in G.nodes()])\n",
    "    plt.title(\"Bipartite Graph Visualization\")\n",
    "    plt.show()\n",
    "\n",
    "def preferential_attachment(d_node,t_node):\n",
    "    d_degree = G.degree(d_node)\n",
    "    t_degree = G.degree(t_node)\n",
    "\n",
    "    pa = d_degree * t_degree\n",
    "    return pa\n",
    "\n",
    "def common_neighbor_jaccard_coef(G, d_node, t_node):\n",
    "    # Get the neighbors of d_node and t_node\n",
    "    d_neighbors = set(G.neighbors(d_node))\n",
    "    t_neighbors = set(G.neighbors(t_node))\n",
    "\n",
    "    # Get the extended neighbors of the d_neighbors and the t_neighbors\n",
    "    d_neighbors_2 = set()\n",
    "    for neighbor in d_neighbors:\n",
    "        d_neighbors_2.update(G.neighbors(neighbor)) \n",
    "        \n",
    "    t_neighbors_2 = set()\n",
    "    for neighbor in t_neighbors:\n",
    "        t_neighbors_2.update(G.neighbors(neighbor))\n",
    "    \n",
    "    # print(\"Second-degree neighbors from d_node:\", d_neighbors_2)\n",
    "    # print(\"Second-degree neighbors from t_node:\", t_neighbors_2)\n",
    "    \n",
    "    # Find the common neighbors, using definition from New J Phys 17 (2015) 113037, where it's the number of nodes in the quadrangles between two seed nodes\n",
    "    common_neighbors_d = d_neighbors_2 & t_neighbors\n",
    "    # print('Common_neighbors_d:', common_neighbors_d)\n",
    "    common_neighbor_count_d = len(common_neighbors_d)\n",
    "    # print(common_neighbor_count_d)\n",
    "    common_neighbors_t = t_neighbors_2 & d_neighbors\n",
    "    # print('Common_neighbors_t:', common_neighbors_t)\n",
    "    common_neighbor_count_t = len(common_neighbors_t)\n",
    "    # print(common_neighbor_count_t)\n",
    "\n",
    "    # assert common_neighbor_count_d == common_neighbor_count_t\n",
    "    # Add up nodes from common_neighbors from d partite, and from t partite\n",
    "    sum_common_neighbors_d_t = len(common_neighbors_d|common_neighbors_t)\n",
    "    total_neighbors = len(d_neighbors|t_neighbors)\n",
    "\n",
    "    # Calculate jaccard coefficient by deviding by the total amount of neighbors from d_node & t_node\n",
    "    jaccard_coefficient_d_t = sum_common_neighbors_d_t / total_neighbors\n",
    "    # print('Total neighbors:', total_neighbors)\n",
    "    print('Jaccard_coefficient:', jaccard_coefficient_d_t)\n",
    "    return sum_common_neighbors_d_t, jaccard_coefficient_d_t\n",
    "\n",
    "# Fuction to calculate precision at top L and AUPR\n",
    "def calculate_metric(df, score_type, l):\n",
    "    # Sort the DataFrame based on the prediction score of the algorithm\n",
    "    sorted_df = df.sort_values(by=score_type, ascending=False)\n",
    "    \n",
    "    # Calculate precision at top L\n",
    "    top_L = sorted_df.head(l)\n",
    "    precision_at_L = top_L['actual outcome'].sum() / l\n",
    "    \n",
    "    # Calculate AUPR\n",
    "    precision, recall, _ = precision_recall_curve(sorted_df['actual outcome'], sorted_df[score_type])\n",
    "    aupr = auc(recall, precision)\n",
    "    \n",
    "    return precision_at_L, aupr   "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 4,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Number of edges in this graph:  75\n",
      "Number of all possible edges in this graph:  105\n",
      "Fold 1:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "  Test set : [('d1', 't1'), ('d1', 't7'), ('d2', 't6'), ('d3', 't2'), ('d3', 't5'), ('d4', 't4'), ('d6', 't6'), ('d7', 't5'), ('d7', 't6'), ('d9', 't1'), ('d10', 't3'), ('d12', 't6'), ('d13', 't1'), ('d13', 't2'), ('d14', 't6')]\n",
      "Fold 2:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "  Test set : [('d2', 't1'), ('d2', 't3'), ('d4', 't2'), ('d5', 't6'), ('d7', 't1'), ('d7', 't2'), ('d7', 't4'), ('d8', 't3'), ('d8', 't4'), ('d9', 't4'), ('d9', 't5'), ('d9', 't7'), ('d11', 't5'), ('d12', 't2'), ('d13', 't6')]\n",
      "Fold 3:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "  Test set : [('d1', 't6'), ('d2', 't2'), ('d2', 't4'), ('d3', 't6'), ('d4', 't3'), ('d4', 't6'), ('d6', 't3'), ('d7', 't7'), ('d8', 't2'), ('d10', 't4'), ('d11', 't1'), ('d11', 't3'), ('d12', 't7'), ('d15', 't3'), ('d15', 't7')]\n",
      "Fold 4:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "  Test set : [('d3', 't3'), ('d4', 't1'), ('d6', 't2'), ('d6', 't4'), ('d6', 't5'), ('d7', 't3'), ('d8', 't7'), ('d9', 't3'), ('d9', 't6'), ('d10', 't1'), ('d11', 't4'), ('d11', 't7'), ('d13', 't7'), ('d14', 't5'), ('d15', 't6')]\n",
      "Fold 5:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "  Test set : [('d1', 't2'), ('d1', 't3'), ('d3', 't7'), ('d5', 't3'), ('d5', 't5'), ('d5', 't7'), ('d6', 't7'), ('d8', 't1'), ('d10', 't5'), ('d10', 't6'), ('d12', 't3'), ('d12', 't4'), ('d13', 't4'), ('d14', 't7'), ('d15', 't2')]\n",
      "Fold 1:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "Fold 2:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "Fold 3:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "Fold 4:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "Fold 5:\n",
      "  Training set size: 60\n",
      "  Test set size: 15\n",
      "B_train_1 number of edges: 60\n",
      "B_train_2 number of edges: 60\n",
      "B_train_3 number of edges: 60\n",
      "B_train_4 number of edges: 60\n",
      "B_train_5 number of edges: 60\n"
     ]
    },
    {
     "data": {
      "image/png": "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",
      "text/plain": [
       "<Figure size 1200x800 with 1 Axes>"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "# Construct link prediction based on various methods, without using community \n",
    "# Input: Given a particular reaction class, determine the likelihood for a particular catalyst to be able to conduct that reaction through topology.\n",
    "\n",
    "# Training-set: remove some edges\n",
    "# Use link prediction algorithm to guess if there's a link\n",
    "# compare link prediction algorithm and edges that were removed, use AUC to evaluate and get average AUC across all cross validation folds\n",
    "\n",
    "# Define cluster for each group of d_nodes here:\n",
    "clusters = {\n",
    "    'c1': ['d1', 'd8', 'd3'],\n",
    "    'c2': ['d4', 'd15'],\n",
    "    'c3': ['d12', 'd7', 'd2', 'd9', 'd10'],\n",
    "    'c4': ['d11', 'd6'],\n",
    "    'c5': ['d13','d14','d5']\n",
    "}\n",
    "# Create a bipartite graph (Replace with bipartite graph from experiments)\n",
    "B = nx.Graph()\n",
    "\n",
    "# Add 15 nodes to the first partite set\n",
    "d_nodes = [f'd{i}' for i in range(1, 16)]\n",
    "B.add_nodes_from(d_nodes, bipartite=0)\n",
    "\n",
    "# Add 7 nodes to the second partite set\n",
    "t_nodes = [f't{i}' for i in range(1, 8)]\n",
    "B.add_nodes_from(t_nodes, bipartite=1)\n",
    "\n",
    "# Add edges between the two sets to form a complete bipartite graph\n",
    "edges = [(d, t) for d in d_nodes for t in t_nodes]\n",
    "B.add_edges_from(edges)\n",
    "\n",
    "# Remove 30 random edges to create the missing links\n",
    "random.seed(42)\n",
    "edges_to_remove = random.sample(edges, 30)\n",
    "B.remove_edges_from(edges_to_remove)\n",
    "\n",
    "# Get the list of remaining edges\n",
    "remaining_edges = list(B.edges())\n",
    "\n",
    "# Get all possible pairs of d nodes and t nodes in the graph\n",
    "all_pairs = list(itertools.product(d_nodes, t_nodes))\n",
    "\n",
    "#Store number of d nodes, number of t nodes\n",
    "num_d_nodes = len(d_nodes)\n",
    "num_t_nodes = len(t_nodes)\n",
    "\n",
    "# Identify number of missing edges from original graph:\n",
    "# Get the pairs that have edges in training set graph\n",
    "total_edges_set = set(B.edges())\n",
    "\n",
    "print(\"Number of edges in this graph: \", len(total_edges_set))\n",
    "print(\"Number of all possible edges in this graph: \", len(set(all_pairs)))\n",
    "\n",
    "# Find pairs with no edges between them\n",
    "original_no_edge_pairs = set(all_pairs) - total_edges_set\n",
    "original_no_edge_pairs = list(original_no_edge_pairs)\n",
    "    \n",
    "\n",
    "# Define 5-fold cross-validation\n",
    "kf = KFold(n_splits=5, shuffle=True, random_state=42)\n",
    "\n",
    "# Split the edges into 5 folds\n",
    "folds = list(kf.split(remaining_edges))\n",
    "\n",
    "# Create 5-fold cross-validation sets\n",
    "cv_sets = []\n",
    "for train_index, test_index in folds:\n",
    "    train_edges = [remaining_edges[i] for i in train_index]\n",
    "    test_edges = [remaining_edges[i] for i in test_index]\n",
    "    cv_sets.append((train_edges, test_edges))\n",
    "\n",
    "# Create a list to store the new graphs\n",
    "B_train_list = []\n",
    "\n",
    "# Create a dictionary to store the test-edges for each graph \n",
    "dict_test_edges = {}\n",
    "\n",
    "# Verify the cross-validation sets\n",
    "for i, (train_edges, test_edges) in enumerate(cv_sets):\n",
    "    print(f\"Fold {i+1}:\")\n",
    "    print(f\"  Training set size: {len(train_edges)}\")\n",
    "    print(f\"  Test set size: {len(test_edges)}\")\n",
    "    print(f\"  Test set : {test_edges}\")\n",
    "    dict_test_edges[f'test_edge_{i}'] = test_edges\n",
    "\n",
    "# Create new graphs for cross validation by removing test edges for each fold\n",
    "for i, (train_edges, test_edges) in enumerate(cv_sets):\n",
    "    B_train = B.copy()\n",
    "    B_train.remove_edges_from(test_edges)\n",
    "    B_train_list.append(B_train)\n",
    "\n",
    "    # Print the sizes of the training and test sets\n",
    "    print(f\"Fold {i+1}:\")\n",
    "    print(f\"  Training set size: {len(train_edges)}\")\n",
    "    print(f\"  Test set size: {len(test_edges)}\")\n",
    "\n",
    "# Verify the new graphs\n",
    "for i, B_train in enumerate(B_train_list):\n",
    "    print(f\"B_train_{i+1} number of edges: {len(B_train.edges())}\")\n",
    "\n",
    "#Visualize one of the training set graphs\n",
    "visualize_bipartite_graph(B_train_list[0], d_nodes, t_nodes)\n",
    "\n",
    "# Dictionary to store DataFrames for each graph in each cross validation fold\n",
    "df_dict_dt = {}   #store graph for both nodes\n",
    "df_dict_cluster_dt = {} #store graph for cluster and t node\n",
    "df_dict_merged_dt ={} #concated dataframe of both types of graphs"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "metadata": {},
   "outputs": [
    {
     "data": {
      "text/plain": [
       "[('d1', 't1'),\n",
       " ('d1', 't7'),\n",
       " ('d2', 't6'),\n",
       " ('d3', 't2'),\n",
       " ('d3', 't5'),\n",
       " ('d4', 't4'),\n",
       " ('d6', 't6'),\n",
       " ('d7', 't5'),\n",
       " ('d7', 't6'),\n",
       " ('d9', 't1'),\n",
       " ('d10', 't3'),\n",
       " ('d12', 't6'),\n",
       " ('d13', 't1'),\n",
       " ('d13', 't2'),\n",
       " ('d14', 't6')]"
      ]
     },
     "execution_count": 5,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "dict_test_edges['test_edge_0']"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 6,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "d_node:  d13  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d5  t_node:  t4\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d6  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d15  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d12  t_node:  t6\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d1  t_node:  t5\n",
      "Jaccard_coefficient: 0.8888888888888888\n",
      "d_node:  d14  t_node:  t2\n",
      "Jaccard_coefficient: 0.7\n",
      "d_node:  d12  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d14  t_node:  t3\n",
      "Jaccard_coefficient: 0.7857142857142857\n",
      "d_node:  d1  t_node:  t4\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d14  t_node:  t6\n",
      "Jaccard_coefficient: 0.8\n",
      "d_node:  d8  t_node:  t5\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d4  t_node:  t5\n",
      "Jaccard_coefficient: 0.9\n",
      "d_node:  d5  t_node:  t2\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d10  t_node:  t7\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d14  t_node:  t1\n",
      "Jaccard_coefficient: 0.75\n",
      "d_node:  d3  t_node:  t5\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d7  t_node:  t5\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d10  t_node:  t2\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d3  t_node:  t4\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d4  t_node:  t4\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d10  t_node:  t3\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d13  t_node:  t5\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d2  t_node:  t5\n",
      "Jaccard_coefficient: 0.9\n",
      "d_node:  d1  t_node:  t7\n",
      "Jaccard_coefficient: 0.9285714285714286\n",
      "d_node:  d15  t_node:  t5\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d5  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d9  t_node:  t2\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d12  t_node:  t5\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d15  t_node:  t4\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d4  t_node:  t7\n",
      "Jaccard_coefficient: 0.9333333333333333\n",
      "d_node:  d1  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d3  t_node:  t2\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d8  t_node:  t6\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d7  t_node:  t6\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d9  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d2  t_node:  t7\n",
      "Jaccard_coefficient: 0.9333333333333333\n",
      "d_node:  d14  t_node:  t4\n",
      "Jaccard_coefficient: 0.9090909090909091\n",
      "d_node:  d13  t_node:  t2\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d3  t_node:  t1\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d11  t_node:  t2\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d2  t_node:  t6\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d13  t_node:  t3\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d11  t_node:  t6\n",
      "Jaccard_coefficient: 1.0\n",
      "d_node:  d6  t_node:  t6\n",
      "Jaccard_coefficient: 1.0\n",
      "Number of rows in node dataframe:  45\n",
      "Edges in the new graph with weights:\n",
      "(t1, c1, 1)\n",
      "(t1, c2, 1)\n",
      "(t1, c3, 3)\n",
      "(t1, c4, 1)\n",
      "(t2, c1, 2)\n",
      "(t2, c2, 2)\n",
      "(t2, c3, 3)\n",
      "(t2, c4, 1)\n",
      "(t3, c1, 3)\n",
      "(t3, c2, 2)\n",
      "(t3, c3, 4)\n",
      "(t3, c4, 2)\n",
      "(t3, c5, 1)\n",
      "(t4, c1, 1)\n",
      "(t4, c3, 5)\n",
      "(t4, c4, 2)\n",
      "(t4, c5, 1)\n",
      "(t5, c3, 2)\n",
      "(t5, c4, 2)\n",
      "(t5, c5, 2)\n",
      "(t6, c1, 2)\n",
      "(t6, c2, 2)\n",
      "(t6, c3, 2)\n",
      "(t6, c5, 2)\n",
      "(t7, c1, 2)\n",
      "(t7, c2, 1)\n",
      "(t7, c3, 3)\n",
      "(t7, c4, 2)\n",
      "(t7, c5, 3)\n",
      "t1 c1 1\n"
     ]
    },
    {
     "ename": "NameError",
     "evalue": "name 'count' is not defined",
     "output_type": "error",
     "traceback": [
      "\u001b[1;31m---------------------------------------------------------------------------\u001b[0m",
      "\u001b[1;31mNameError\u001b[0m                                 Traceback (most recent call last)",
      "Cell \u001b[1;32mIn[6], line 91\u001b[0m\n\u001b[0;32m     89\u001b[0m \u001b[38;5;28mprint\u001b[39m(t,c, weight)\n\u001b[0;32m     90\u001b[0m \u001b[38;5;28;01mfor\u001b[39;00m d \u001b[38;5;129;01min\u001b[39;00m clusters[c]:\n\u001b[1;32m---> 91\u001b[0m     count \u001b[38;5;241m=\u001b[39m count \u001b[38;5;241m+\u001b[39m \u001b[38;5;241m1\u001b[39m\n\u001b[0;32m     92\u001b[0m     degree_d_cluster \u001b[38;5;241m=\u001b[39m H\u001b[38;5;241m.\u001b[39mdegree(c)\n\u001b[0;32m     93\u001b[0m     degree_t \u001b[38;5;241m=\u001b[39m H\u001b[38;5;241m.\u001b[39mdegree(t)\n",
      "\u001b[1;31mNameError\u001b[0m: name 'count' is not defined"
     ]
    }
   ],
   "source": [
    "# Iterate through every B in B_train_list\n",
    "# Generate dataframe of 'test set' edges for every link prediction algorithm (neighbor/cluster based), assuming cluster is given\n",
    "# Compare prediction trend of test set with the actual test set predictions\n",
    "\n",
    "# Iterate each graph within training set\n",
    "for idx, G in enumerate(B_train_list):\n",
    "# for G in B_train_list:\n",
    "    # Get the pairs that have edges in training set graph\n",
    "    edges_set = set(G.edges())\n",
    "\n",
    "    # print(\"Number of edges in this graph for training set: \", len(edges_set))\n",
    "    # print(\"Number of all possible edges in this graph for training set: \", len(set(all_pairs)))\n",
    "    # Find pairs with no edges between them\n",
    "    training_no_edge_pairs = set(all_pairs) - edges_set\n",
    "    training_no_edge_pairs = list(training_no_edge_pairs)\n",
    "\n",
    "    # Ensure that the missing edges in each training set is equal to the sum of both the number of original missing edges and the additional ones from the test set\n",
    "    assert len(training_no_edge_pairs) == len(original_no_edge_pairs) + len(test_edges)\n",
    "    \n",
    "    data = []\n",
    "    # Calculate common neighbors and Jaccard coefficient for each pair without edges\n",
    "    for d_node, t_node in training_no_edge_pairs:\n",
    "        print('d_node: ', d_node, ' t_node: ', t_node)\n",
    "        num_common_neighbor, jaccard_coefficient = common_neighbor_jaccard_coef(G, d_node, t_node)\n",
    "        preferential_d_t = preferential_attachment(d_node,t_node)\n",
    "        data.append({\n",
    "            'd_node': d_node,\n",
    "            't_node': t_node,\n",
    "            'num_common_neighbor': num_common_neighbor,\n",
    "            'jaccard_coefficient': jaccard_coefficient,\n",
    "            'preferential_attach' : preferential_d_t\n",
    "        })\n",
    "    \n",
    "    # Create the DataFrame of link prediction algorithm containing each d t pair that has missing edges, store into dctionary of dataframes\n",
    "    df_neighbor = pd.DataFrame(data)\n",
    "    df_dict_dt[f'graph_{idx}'] = df_neighbor\n",
    "    print('Number of rows in node dataframe: ', df_dict_dt[f'graph_{idx}'].shape[0])\n",
    "\n",
    "    # Step 3: Create a new graph connecting t_nodes to clusters\n",
    "    H = nx.Graph()\n",
    "    H_zero = nx.Graph() #Add a H_zero graph to store data for links where weight = 0 \n",
    "    # Add t nodes to the new graph\n",
    "    H.add_nodes_from(t_nodes, bipartite=1)\n",
    "    H_zero.add_nodes_from(t_nodes, bipartite=1)\n",
    "    # Add cluster nodes to the new graph\n",
    "    H.add_nodes_from(clusters.keys(), bipartite=0)\n",
    "    H_zero.add_nodes_from(clusters.keys(), bipartite=0)\n",
    "\n",
    "    num_clusters = len(clusters)\n",
    "    \n",
    "    # Step 4: Calculate the weights for links between clusters and t_nodes\n",
    "    # Initialize a dictionary to store weights for cluster-t_node links\n",
    "    cluster_t_weights = {cluster: {t: 0 for t in t_nodes} for cluster in clusters}\n",
    "    \n",
    "    # Create list for links with zero weight\n",
    "    cluster_t_zero_weights = {cluster: {t: 0 for t in t_nodes} for cluster in clusters}    \n",
    "    \n",
    "    # Iterate over the original edges and update the weights for the new graph\n",
    "    for d, t in G.edges():\n",
    "        for cluster, d_list in clusters.items():\n",
    "            if d in d_list:\n",
    "                cluster_t_weights[cluster][t] += 1\n",
    "    \n",
    "    # Add weighted edges to the new graph\n",
    "    for cluster, t_weights in cluster_t_weights.items():\n",
    "        for t, w in t_weights.items():\n",
    "            if w > 0:  # Only add edges with non-zero weight\n",
    "                H.add_edge(cluster, t, weight=w)    #Missing some t, cluster combinations since there is no link.\n",
    "            elif w == 0:\n",
    "                H_zero.add_edge(cluster, t, weight=w)\n",
    "    \n",
    "    # Print the new graph edges with weights\n",
    "    print(\"Edges in the new graph with weights:\")\n",
    "    \n",
    "    for (u, v, wt) in H.edges(data='weight'):\n",
    "        print(f\"({u}, {v}, {wt})\")\n",
    "        \n",
    "    cluster_edge_weights_list = list(H.edges(data='weight'))  # list of weights for c-t edges\n",
    "    cluster_edge_zero_weights_list = list(H_zero.edges(data='weight')) # list of missing links for c-t combinations\n",
    "\n",
    "    #Make sure number of cluster x number of t nodes equal to the number of all possible edges between cluster and t nodes\n",
    "    assert len(clusters) * num_t_nodes == len(cluster_edge_weights_list)+len(cluster_edge_zero_weights_list) \n",
    "    \n",
    "    # Create a list to store the data\n",
    "    data_cluster = []\n",
    "    count = 0\n",
    "    # Iterate through the clusters and their connections to t_nodes\n",
    "    for t, c, weight in cluster_edge_weights_list:\n",
    "        print(t,c, weight)\n",
    "        for d in clusters[c]:\n",
    "            count = count + 1\n",
    "            degree_d_cluster = H.degree(c)\n",
    "            degree_t = H.degree(t)\n",
    "            data_cluster.append((d, \n",
    "                                 t, \n",
    "                                 weight, \n",
    "                                 len(clusters[c]),\n",
    "                                 weight/len(clusters[c]),\n",
    "                                 degree_d_cluster,\n",
    "                                 degree_t,\n",
    "                                 degree_d_cluster*degree_t))\n",
    "    # Add the t_node cluster pair with no weights in to the dataframe as well, so we can get all 105 combinations of d and t\n",
    "    for t, c, weight in cluster_edge_zero_weights_list:\n",
    "        for d in clusters[c]:\n",
    "            count = count + 1\n",
    "            degree_d_cluster = H.degree(c)\n",
    "            degree_t = H.degree(t)\n",
    "            data_cluster.append((d, \n",
    "                                 t, \n",
    "                                 weight, \n",
    "                                 len(clusters[c]),\n",
    "                                 weight/len(clusters[c]),\n",
    "                                 degree_d_cluster,\n",
    "                                 degree_t,\n",
    "                                 degree_d_cluster*degree_t))\n",
    "            \n",
    "    # If t, cluster has no link, then we still need to record the d and t involved, where weight = 0, number of d in cluster is based on c\n",
    "    # degree of cluster is equal to cluster, # jaccard coefficient is 0, degree_t is based on t preferential_attachment is calculated properly\n",
    "    \n",
    "    # Create the dataframe for cluster-t \n",
    "    cluster_df = pd.DataFrame(data_cluster, columns=['d_node', \n",
    "                                                     't_node', \n",
    "                                                     'common_neighbor_tc',\n",
    "                                                     'number of d in cluster',\n",
    "                                                     'jaccard_coeff_tc',\n",
    "                                                     'degree_d_cluster',\n",
    "                                                     'degree_t',\n",
    "                                                     'preferential_attachment_cluster'])\n",
    "    \n",
    "    # print('Number of rows in cluster dataframe: ', cluster_df.shape[0])  #should have 105?\n",
    "\n",
    "    # Make sure that the number of rows for the cluster dataframe is equal to the total possible combinations of all node connections\n",
    "    assert cluster_df.shape[0] == len(set(all_pairs))\n",
    "\n",
    "    # Create a DataFrame from training_no_edge_pairs\n",
    "    training_no_edge_pairs_df = pd.DataFrame(training_no_edge_pairs, columns=['d_node', 't_node'])\n",
    "    print('Number of rows in no edge pair dataframe: ', training_no_edge_pairs_df.shape[0])\n",
    "    \n",
    "    # Merge the cluster_df with training_no_edge_pairs_df on d_node and t_node\n",
    "    filtered_cluster_df = pd.merge(cluster_df, training_no_edge_pairs_df, on=['d_node', 't_node'])\n",
    "    \n",
    "    # Store dataframe into a dictionary of dataframes\n",
    "    df_dict_cluster_dt[f'graph_{idx}'] = filtered_cluster_df\n",
    "\n",
    "    df_dict_merged_dt[f'graph_{idx}'] = pd.merge(df_dict_dt[f'graph_{idx}'], df_dict_cluster_dt[f'graph_{idx}'], on=['d_node', 't_node'], how='inner')\n",
    "    # print('Number of rows in merged dataframe: ', df_dict_merged_dt[f'graph_{idx}'].shape[0])\n",
    "\n",
    "    # Make sure that the number of rows after merging is equal to the number of missing pairs from the training graph \n",
    "    assert df_dict_merged_dt[f'graph_{idx}'].shape[0] == len(training_no_edge_pairs)\n",
    "\n",
    "# Check graph_0 within both dictionary\n",
    "# print('Dataframe_neighbors: ',df_dict_dt['graph_0'])\n",
    "# print('Dataframe_clusters: ',df_dict_cluster_dt['graph_0'])  \n",
    "# print('Merged dataframe: ',df_dict_merged_dt['graph_0']) \n",
    "column_list = df_dict_merged_dt['graph_0'].columns.tolist() \n",
    "print(column_list)\n",
    "        "
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": [
    "df_dict_dt['graph_0'].to_csv('dataframe_neighbors.csv', index=False)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 29,
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "List of test edges: {('d9', 't1'), ('d1', 't7'), ('d10', 't3'), ('d6', 't6'), ('d7', 't6'), ('d14', 't6'), ('d13', 't1'), ('d3', 't5'), ('d7', 't5'), ('d4', 't4'), ('d1', 't1'), ('d3', 't2'), ('d12', 't6'), ('d2', 't6'), ('d13', 't2')}\n",
      "   d_node t_node  num_common_neighbor  jaccard_coefficient  \\\n",
      "0     d12     t5                  0.4             1.000000   \n",
      "1      d1     t7                  0.7             0.761905   \n",
      "2     d14     t4                  0.4             0.696970   \n",
      "3      d6     t6                  0.7             1.000000   \n",
      "4     d15     t5                  0.4             1.000000   \n",
      "5     d14     t6                  0.2             0.333333   \n",
      "6     d13     t1                  0.3             1.000000   \n",
      "7      d7     t5                  0.5             1.000000   \n",
      "8      d1     t1                  0.3             1.000000   \n",
      "9     d12     t6                  0.6             1.000000   \n",
      "10     d2     t5                  0.3             0.666667   \n",
      "11     d4     t5                  0.3             0.666667   \n",
      "12    d11     t2                  0.7             1.000000   \n",
      "13     d8     t5                  0.5             1.000000   \n",
      "14    d15     t4                  0.7             1.000000   \n",
      "15     d5     t2                  0.6             1.000000   \n",
      "16     d7     t6                  0.7             1.000000   \n",
      "17     d3     t5                  0.3             1.000000   \n",
      "18     d6     t1                  0.5             1.000000   \n",
      "19    d11     t6                  0.7             1.000000   \n",
      "20    d14     t1                  0.0             0.166667   \n",
      "21     d9     t2                  0.7             1.000000   \n",
      "22     d4     t4                  0.7             1.000000   \n",
      "23    d14     t3                  0.5             0.285714   \n",
      "24     d5     t4                  0.7             1.000000   \n",
      "25     d3     t2                  0.5             1.000000   \n",
      "26     d2     t6                  0.6             1.000000   \n",
      "27    d12     t1                  0.4             1.000000   \n",
      "28     d8     t6                  0.7             1.000000   \n",
      "29     d3     t4                  0.6             1.000000   \n",
      "30    d10     t3                  1.0             1.000000   \n",
      "31    d15     t1                  0.4             1.000000   \n",
      "32     d4     t7                  0.8             0.777778   \n",
      "33     d2     t7                  0.8             0.777778   \n",
      "34    d10     t2                  0.6             1.000000   \n",
      "35    d13     t5                  0.3             1.000000   \n",
      "36    d13     t3                  0.9             1.000000   \n",
      "37     d1     t5                  0.2             0.629630   \n",
      "38    d13     t2                  0.5             1.000000   \n",
      "39     d5     t1                  0.4             1.000000   \n",
      "40     d9     t1                  0.5             1.000000   \n",
      "41     d1     t4                  0.6             1.000000   \n",
      "42     d3     t1                  0.3             1.000000   \n",
      "43    d10     t7                  0.9             1.000000   \n",
      "44    d14     t2                  0.1             0.000000   \n",
      "\n",
      "    preferential_attach  common_neighbor_tc  number of d in cluster  \\\n",
      "0              0.333333                0.50                       5   \n",
      "1              0.583333                0.50                       3   \n",
      "2              0.166667                0.25                       3   \n",
      "3              0.777778                0.00                       2   \n",
      "4              0.333333                0.00                       2   \n",
      "5              0.111111                0.50                       3   \n",
      "6              0.166667                0.00                       3   \n",
      "7              0.500000                0.50                       5   \n",
      "8              0.166667                0.25                       3   \n",
      "9              0.555556                0.50                       5   \n",
      "10             0.333333                0.50                       5   \n",
      "11             0.333333                0.00                       2   \n",
      "12             0.777778                0.25                       2   \n",
      "13             0.500000                0.00                       3   \n",
      "14             0.666667                0.00                       2   \n",
      "15             0.555556                0.00                       3   \n",
      "16             0.777778                0.50                       5   \n",
      "17             0.166667                0.00                       3   \n",
      "18             0.500000                0.25                       2   \n",
      "19             0.777778                0.00                       2   \n",
      "20             0.000000                0.00                       3   \n",
      "21             0.777778                0.75                       5   \n",
      "22             0.666667                0.00                       2   \n",
      "23             0.333333                0.25                       3   \n",
      "24             0.666667                0.25                       3   \n",
      "25             0.333333                0.50                       3   \n",
      "26             0.555556                0.50                       5   \n",
      "27             0.333333                0.75                       5   \n",
      "28             0.777778                0.50                       3   \n",
      "29             0.416667                0.25                       3   \n",
      "30             1.000000                1.00                       5   \n",
      "31             0.333333                0.25                       2   \n",
      "32             0.888889                0.25                       2   \n",
      "33             0.888889                0.75                       5   \n",
      "34             0.555556                0.75                       5   \n",
      "35             0.166667                0.50                       3   \n",
      "36             0.666667                0.25                       3   \n",
      "37             0.166667                0.00                       3   \n",
      "38             0.333333                0.00                       3   \n",
      "39             0.333333                0.00                       3   \n",
      "40             0.500000                0.75                       5   \n",
      "41             0.416667                0.25                       3   \n",
      "42             0.166667                0.25                       3   \n",
      "43             0.888889                0.75                       5   \n",
      "44             0.111111                0.00                       3   \n",
      "\n",
      "    jaccard_coeff_tc  degree_d_cluster  degree_t  \\\n",
      "0           0.500000                 7         3   \n",
      "1           0.833333                 6         5   \n",
      "2           0.416667                 5         4   \n",
      "3           0.000000                 6         4   \n",
      "4           0.000000                 5         3   \n",
      "5           0.833333                 5         4   \n",
      "6           0.000000                 5         4   \n",
      "7           0.500000                 7         3   \n",
      "8           0.416667                 6         4   \n",
      "9           0.500000                 7         4   \n",
      "10          0.500000                 7         3   \n",
      "11          0.000000                 5         3   \n",
      "12          0.625000                 6         4   \n",
      "13          0.000000                 6         3   \n",
      "14          0.000000                 5         4   \n",
      "15          0.000000                 5         4   \n",
      "16          0.500000                 7         4   \n",
      "17          0.000000                 6         3   \n",
      "18          0.625000                 6         4   \n",
      "19          0.000000                 6         4   \n",
      "20          0.000000                 5         4   \n",
      "21          0.750000                 7         4   \n",
      "22          0.000000                 5         4   \n",
      "23          0.416667                 5         5   \n",
      "24          0.416667                 5         4   \n",
      "25          0.833333                 6         4   \n",
      "26          0.500000                 7         4   \n",
      "27          0.750000                 7         4   \n",
      "28          0.833333                 6         4   \n",
      "29          0.416667                 6         4   \n",
      "30          1.000000                 7         5   \n",
      "31          0.625000                 5         4   \n",
      "32          0.625000                 5         5   \n",
      "33          0.750000                 7         5   \n",
      "34          0.750000                 7         4   \n",
      "35          0.833333                 5         3   \n",
      "36          0.416667                 5         5   \n",
      "37          0.000000                 6         3   \n",
      "38          0.000000                 5         4   \n",
      "39          0.000000                 5         4   \n",
      "40          0.750000                 7         4   \n",
      "41          0.416667                 6         4   \n",
      "42          0.416667                 6         4   \n",
      "43          0.750000                 7         5   \n",
      "44          0.000000                 5         4   \n",
      "\n",
      "    preferential_attachment_cluster  actual outcome  \n",
      "0                              0.30               0  \n",
      "1                              0.75               1  \n",
      "2                              0.25               0  \n",
      "3                              0.45               1  \n",
      "4                              0.00               0  \n",
      "5                              0.25               1  \n",
      "6                              0.25               1  \n",
      "7                              0.30               1  \n",
      "8                              0.45               1  \n",
      "9                              0.65               1  \n",
      "10                             0.30               0  \n",
      "11                             0.00               0  \n",
      "12                             0.45               0  \n",
      "13                             0.15               0  \n",
      "14                             0.25               0  \n",
      "15                             0.25               0  \n",
      "16                             0.65               1  \n",
      "17                             0.15               1  \n",
      "18                             0.45               0  \n",
      "19                             0.45               0  \n",
      "20                             0.25               0  \n",
      "21                             0.65               0  \n",
      "22                             0.25               1  \n",
      "23                             0.50               0  \n",
      "24                             0.25               0  \n",
      "25                             0.45               1  \n",
      "26                             0.65               1  \n",
      "27                             0.65               0  \n",
      "28                             0.45               0  \n",
      "29                             0.45               0  \n",
      "30                             1.00               1  \n",
      "31                             0.25               0  \n",
      "32                             0.50               0  \n",
      "33                             1.00               0  \n",
      "34                             0.65               0  \n",
      "35                             0.00               0  \n",
      "36                             0.50               0  \n",
      "37                             0.15               0  \n",
      "38                             0.25               1  \n",
      "39                             0.25               0  \n",
      "40                             0.65               1  \n",
      "41                             0.45               0  \n",
      "42                             0.45               0  \n",
      "43                             1.00               0  \n",
      "44                             0.25               0  \n",
      "         Link Prediction Algorithm  Precision      AUPR\n",
      "0              num_common_neighbor   0.333333  0.368557\n",
      "1              jaccard_coefficient   0.333333  0.646956\n",
      "2              preferential_attach   0.333333  0.373113\n",
      "3               common_neighbor_tc   0.466667  0.431746\n",
      "4                 jaccard_coeff_tc   0.333333  0.501515\n",
      "5  preferential_attachment_cluster   0.466667  0.437936\n",
      "List of test edges: {('d2', 't1'), ('d7', 't4'), ('d9', 't5'), ('d2', 't3'), ('d13', 't6'), ('d4', 't2'), ('d7', 't1'), ('d8', 't3'), ('d9', 't7'), ('d12', 't2'), ('d9', 't4'), ('d5', 't6'), ('d11', 't5'), ('d8', 't4'), ('d7', 't2')}\n",
      "   d_node t_node  num_common_neighbor  jaccard_coefficient  \\\n",
      "0     d12     t5                0.250             1.000000   \n",
      "1     d14     t4                0.250             1.000000   \n",
      "2     d15     t5                0.250             1.000000   \n",
      "3     d12     t2                0.375             1.000000   \n",
      "4     d11     t5                0.250             1.000000   \n",
      "5      d2     t5                0.000             0.000000   \n",
      "6      d7     t2                0.375             1.000000   \n",
      "7      d7     t4                0.375             1.000000   \n",
      "8     d11     t2                0.375             1.000000   \n",
      "9      d8     t5                0.125             1.000000   \n",
      "10     d9     t5                0.125             1.000000   \n",
      "11    d15     t4                0.375             1.000000   \n",
      "12     d4     t5                0.250             1.000000   \n",
      "13     d2     t3                0.625             0.357143   \n",
      "14     d5     t2                0.125             0.100000   \n",
      "15     d4     t2                0.375             1.000000   \n",
      "16     d6     t1                0.625             1.000000   \n",
      "17     d8     t3                0.750             1.000000   \n",
      "18    d14     t1                0.250             1.000000   \n",
      "19     d9     t2                0.250             1.000000   \n",
      "20    d11     t6                0.875             1.000000   \n",
      "21    d14     t3                0.750             1.000000   \n",
      "22     d5     t4                0.125             0.100000   \n",
      "23     d9     t4                0.250             1.000000   \n",
      "24     d5     t6                0.625             0.357143   \n",
      "25     d8     t4                0.250             1.000000   \n",
      "26    d12     t1                0.375             1.000000   \n",
      "27     d8     t6                0.750             1.000000   \n",
      "28     d3     t4                0.500             1.000000   \n",
      "29    d15     t1                0.375             1.000000   \n",
      "30     d4     t7                0.875             1.000000   \n",
      "31     d2     t7                0.625             0.357143   \n",
      "32    d10     t2                0.500             1.000000   \n",
      "33     d9     t7                0.750             1.000000   \n",
      "34    d13     t5                0.250             1.000000   \n",
      "35     d7     t1                0.375             1.000000   \n",
      "36    d13     t3                0.875             1.000000   \n",
      "37     d1     t5                0.375             1.000000   \n",
      "38     d5     t1                0.250             1.000000   \n",
      "39     d2     t1                0.250             1.000000   \n",
      "40    d13     t6                0.875             1.000000   \n",
      "41     d1     t4                0.500             1.000000   \n",
      "42     d3     t1                0.500             1.000000   \n",
      "43    d10     t7                1.000             1.000000   \n",
      "44    d14     t2                0.250             1.000000   \n",
      "\n",
      "    preferential_attach  common_neighbor_tc  number of d in cluster  \\\n",
      "0              0.162162                0.50                       5   \n",
      "1              0.081081                0.25                       3   \n",
      "2              0.162162                0.00                       2   \n",
      "3              0.270270                0.25                       5   \n",
      "4              0.162162                0.25                       2   \n",
      "5              0.000000                0.50                       5   \n",
      "6              0.270270                0.25                       5   \n",
      "7              0.270270                0.75                       5   \n",
      "8              0.270270                0.25                       2   \n",
      "9              0.000000                0.25                       3   \n",
      "10             0.000000                0.50                       5   \n",
      "11             0.270270                0.25                       2   \n",
      "12             0.162162                0.00                       2   \n",
      "13             0.405405                1.00                       5   \n",
      "14             0.081081                0.25                       3   \n",
      "15             0.270270                0.25                       2   \n",
      "16             0.648649                0.25                       2   \n",
      "17             0.405405                0.50                       3   \n",
      "18             0.081081                0.25                       3   \n",
      "19             0.081081                0.25                       5   \n",
      "20             0.702703                0.25                       2   \n",
      "21             0.405405                0.25                       3   \n",
      "22             0.081081                0.25                       3   \n",
      "23             0.081081                0.75                       5   \n",
      "24             0.405405                0.25                       3   \n",
      "25             0.081081                0.00                       3   \n",
      "26             0.270270                0.50                       5   \n",
      "27             0.405405                0.50                       3   \n",
      "28             0.459459                0.00                       3   \n",
      "29             0.270270                0.25                       2   \n",
      "30             0.702703                0.25                       2   \n",
      "31             0.405405                0.50                       5   \n",
      "32             0.459459                0.25                       5   \n",
      "33             0.405405                0.50                       5   \n",
      "34             0.162162                0.50                       3   \n",
      "35             0.270270                0.50                       5   \n",
      "36             0.702703                0.25                       3   \n",
      "37             0.324324                0.25                       3   \n",
      "38             0.081081                0.25                       3   \n",
      "39             0.081081                0.50                       5   \n",
      "40             0.702703                0.25                       3   \n",
      "41             0.459459                0.00                       3   \n",
      "42             0.459459                0.50                       3   \n",
      "43             1.000000                0.50                       5   \n",
      "44             0.081081                0.25                       3   \n",
      "\n",
      "    jaccard_coeff_tc  degree_d_cluster  degree_t  \\\n",
      "0           0.500000                 7         4   \n",
      "1           0.416667                 7         4   \n",
      "2           0.000000                 6         4   \n",
      "3           0.250000                 7         5   \n",
      "4           0.625000                 7         4   \n",
      "5           0.500000                 7         4   \n",
      "6           0.250000                 7         5   \n",
      "7           0.750000                 7         4   \n",
      "8           0.625000                 7         5   \n",
      "9           0.416667                 6         4   \n",
      "10          0.500000                 7         4   \n",
      "11          0.625000                 6         4   \n",
      "12          0.000000                 6         4   \n",
      "13          1.000000                 7         5   \n",
      "14          0.416667                 7         5   \n",
      "15          0.625000                 6         5   \n",
      "16          0.625000                 7         5   \n",
      "17          0.833333                 6         5   \n",
      "18          0.416667                 7         5   \n",
      "19          0.250000                 7         5   \n",
      "20          0.625000                 7         5   \n",
      "21          0.416667                 7         5   \n",
      "22          0.416667                 7         4   \n",
      "23          0.750000                 7         4   \n",
      "24          0.416667                 7         5   \n",
      "25          0.000000                 6         4   \n",
      "26          0.500000                 7         5   \n",
      "27          0.833333                 6         5   \n",
      "28          0.000000                 6         4   \n",
      "29          0.625000                 6         5   \n",
      "30          0.625000                 6         5   \n",
      "31          0.500000                 7         5   \n",
      "32          0.250000                 7         5   \n",
      "33          0.500000                 7         5   \n",
      "34          0.833333                 7         4   \n",
      "35          0.500000                 7         5   \n",
      "36          0.416667                 7         5   \n",
      "37          0.416667                 6         4   \n",
      "38          0.416667                 7         5   \n",
      "39          0.500000                 7         5   \n",
      "40          0.416667                 7         5   \n",
      "41          0.000000                 6         4   \n",
      "42          0.833333                 6         5   \n",
      "43          0.500000                 7         5   \n",
      "44          0.416667                 7         5   \n",
      "\n",
      "    preferential_attachment_cluster  actual outcome  \n",
      "0                          0.363636               0  \n",
      "1                          0.363636               0  \n",
      "2                          0.000000               0  \n",
      "3                          1.000000               1  \n",
      "4                          0.363636               1  \n",
      "5                          0.363636               0  \n",
      "6                          1.000000               1  \n",
      "7                          0.363636               1  \n",
      "8                          1.000000               0  \n",
      "9                          0.000000               0  \n",
      "10                         0.363636               1  \n",
      "11                         0.000000               0  \n",
      "12                         0.000000               0  \n",
      "13                         1.000000               1  \n",
      "14                         1.000000               0  \n",
      "15                         0.545455               1  \n",
      "16                         1.000000               0  \n",
      "17                         0.545455               1  \n",
      "18                         1.000000               0  \n",
      "19                         1.000000               0  \n",
      "20                         1.000000               0  \n",
      "21                         1.000000               0  \n",
      "22                         0.363636               0  \n",
      "23                         0.363636               1  \n",
      "24                         1.000000               1  \n",
      "25                         0.000000               1  \n",
      "26                         1.000000               0  \n",
      "27                         0.545455               0  \n",
      "28                         0.000000               0  \n",
      "29                         0.545455               0  \n",
      "30                         0.545455               0  \n",
      "31                         1.000000               0  \n",
      "32                         1.000000               0  \n",
      "33                         1.000000               1  \n",
      "34                         0.363636               0  \n",
      "35                         1.000000               1  \n",
      "36                         1.000000               0  \n",
      "37                         0.000000               0  \n",
      "38                         1.000000               0  \n",
      "39                         1.000000               1  \n",
      "40                         1.000000               1  \n",
      "41                         0.000000               0  \n",
      "42                         0.545455               0  \n",
      "43                         1.000000               0  \n",
      "44                         1.000000               0  \n",
      "         Link Prediction Algorithm  Precision      AUPR\n",
      "0              num_common_neighbor   0.333333  0.319913\n",
      "1              jaccard_coefficient   0.200000  0.623810\n",
      "2              preferential_attach   0.200000  0.280975\n",
      "3               common_neighbor_tc   0.466667  0.642778\n",
      "4                 jaccard_coeff_tc   0.400000  0.470079\n",
      "5  preferential_attachment_cluster   0.266667  0.533482\n",
      "List of test edges: {('d4', 't3'), ('d12', 't7'), ('d3', 't6'), ('d2', 't2'), ('d8', 't2'), ('d15', 't3'), ('d1', 't6'), ('d2', 't4'), ('d6', 't3'), ('d7', 't7'), ('d10', 't4'), ('d11', 't1'), ('d4', 't6'), ('d11', 't3'), ('d15', 't7')}\n",
      "   d_node t_node  num_common_neighbor  jaccard_coefficient  \\\n",
      "0     d12     t5             0.571429             1.000000   \n",
      "1     d14     t4             0.285714             0.545455   \n",
      "2     d15     t5             0.142857             0.500000   \n",
      "3     d15     t3             0.285714             0.545455   \n",
      "4     d11     t3             0.428571             0.583333   \n",
      "5      d2     t5             0.285714             0.545455   \n",
      "6      d4     t5             0.142857             0.090909   \n",
      "7     d11     t2             0.285714             0.545455   \n",
      "8      d8     t5             0.571429             1.000000   \n",
      "9     d15     t4             0.000000             0.000000   \n",
      "10     d4     t3             0.428571             0.583333   \n",
      "11     d5     t2             0.428571             0.583333   \n",
      "12    d12     t7             0.714286             1.000000   \n",
      "13     d6     t1             0.714286             1.000000   \n",
      "14     d2     t2             0.428571             1.000000   \n",
      "15     d8     t2             0.428571             0.583333   \n",
      "16    d14     t1             0.285714             0.545455   \n",
      "17     d9     t2             0.857143             1.000000   \n",
      "18    d11     t6             0.428571             0.230769   \n",
      "19     d2     t4             0.285714             0.545455   \n",
      "20    d14     t3             0.571429             1.000000   \n",
      "21     d5     t4             0.428571             0.583333   \n",
      "22     d7     t7             1.000000             1.000000   \n",
      "23     d4     t6             0.428571             0.230769   \n",
      "24    d15     t7             0.142857             0.090909   \n",
      "25    d12     t1             0.571429             1.000000   \n",
      "26     d8     t6             0.714286             0.642857   \n",
      "27     d3     t4             0.571429             1.000000   \n",
      "28    d15     t1             0.142857             0.500000   \n",
      "29     d4     t7             0.285714             0.166667   \n",
      "30     d2     t7             0.428571             0.583333   \n",
      "31     d3     t6             0.857143             1.000000   \n",
      "32    d10     t2             0.571429             1.000000   \n",
      "33    d13     t5             0.714286             1.000000   \n",
      "34    d10     t4             0.571429             1.000000   \n",
      "35    d11     t1             0.285714             0.545455   \n",
      "36    d13     t3             0.857143             1.000000   \n",
      "37     d1     t5             0.571429             1.000000   \n",
      "38     d5     t1             0.428571             0.583333   \n",
      "39     d1     t4             0.571429             1.000000   \n",
      "40     d3     t1             0.571429             1.000000   \n",
      "41    d10     t7             0.714286             1.000000   \n",
      "42     d1     t6             0.857143             1.000000   \n",
      "43     d6     t3             0.857143             1.000000   \n",
      "44    d14     t2             0.285714             0.545455   \n",
      "\n",
      "    preferential_attach  common_neighbor_tc  number of d in cluster  \\\n",
      "0              0.421053                0.75                       5   \n",
      "1              0.210526                0.25                       3   \n",
      "2              0.000000                0.00                       2   \n",
      "3              0.052632                0.00                       2   \n",
      "4              0.289474                0.00                       2   \n",
      "5              0.210526                0.75                       5   \n",
      "6              0.210526                0.00                       2   \n",
      "7              0.210526                0.25                       2   \n",
      "8              0.421053                0.25                       3   \n",
      "9              0.000000                0.25                       2   \n",
      "10             0.289474                0.00                       2   \n",
      "11             0.421053                0.25                       3   \n",
      "12             0.526316                0.25                       5   \n",
      "13             0.631579                0.00                       2   \n",
      "14             0.210526                0.50                       5   \n",
      "15             0.421053                0.50                       3   \n",
      "16             0.210526                0.25                       3   \n",
      "17             0.842105                0.50                       5   \n",
      "18             0.368421                0.25                       2   \n",
      "19             0.210526                0.75                       5   \n",
      "20             0.289474                0.25                       3   \n",
      "21             0.421053                0.25                       3   \n",
      "22             1.000000                0.25                       5   \n",
      "23             0.368421                0.25                       2   \n",
      "24             0.052632                0.00                       2   \n",
      "25             0.421053                1.00                       5   \n",
      "26             0.631579                0.00                       3   \n",
      "27             0.421053                0.25                       3   \n",
      "28             0.000000                0.25                       2   \n",
      "29             0.289474                0.00                       2   \n",
      "30             0.289474                0.25                       5   \n",
      "31             0.631579                0.00                       3   \n",
      "32             0.421053                0.50                       5   \n",
      "33             0.631579                0.50                       3   \n",
      "34             0.421053                0.75                       5   \n",
      "35             0.210526                0.00                       2   \n",
      "36             0.763158                0.25                       3   \n",
      "37             0.421053                0.25                       3   \n",
      "38             0.421053                0.25                       3   \n",
      "39             0.421053                0.25                       3   \n",
      "40             0.421053                0.50                       3   \n",
      "41             0.526316                0.25                       5   \n",
      "42             0.631579                0.00                       3   \n",
      "43             0.763158                0.00                       2   \n",
      "44             0.210526                0.25                       3   \n",
      "\n",
      "    jaccard_coeff_tc  degree_d_cluster  degree_t  \\\n",
      "0           0.750000                 7         4   \n",
      "1           0.416667                 7         5   \n",
      "2           0.000000                 4         4   \n",
      "3           0.000000                 4         3   \n",
      "4           0.000000                 5         3   \n",
      "5           0.750000                 7         4   \n",
      "6           0.000000                 4         4   \n",
      "7           0.625000                 5         5   \n",
      "8           0.416667                 6         4   \n",
      "9           0.625000                 4         5   \n",
      "10          0.000000                 4         3   \n",
      "11          0.416667                 7         5   \n",
      "12          0.250000                 7         4   \n",
      "13          0.000000                 5         4   \n",
      "14          0.500000                 7         5   \n",
      "15          0.833333                 6         5   \n",
      "16          0.416667                 7         4   \n",
      "17          0.500000                 7         5   \n",
      "18          0.625000                 5         4   \n",
      "19          0.750000                 7         5   \n",
      "20          0.416667                 7         3   \n",
      "21          0.416667                 7         5   \n",
      "22          0.250000                 7         4   \n",
      "23          0.625000                 4         4   \n",
      "24          0.000000                 4         4   \n",
      "25          1.000000                 7         4   \n",
      "26          0.000000                 6         4   \n",
      "27          0.416667                 6         5   \n",
      "28          0.625000                 4         4   \n",
      "29          0.000000                 4         4   \n",
      "30          0.250000                 7         4   \n",
      "31          0.000000                 6         4   \n",
      "32          0.500000                 7         5   \n",
      "33          0.833333                 7         4   \n",
      "34          0.750000                 7         5   \n",
      "35          0.000000                 5         4   \n",
      "36          0.416667                 7         3   \n",
      "37          0.416667                 6         4   \n",
      "38          0.416667                 7         4   \n",
      "39          0.416667                 6         5   \n",
      "40          0.833333                 6         4   \n",
      "41          0.250000                 7         4   \n",
      "42          0.000000                 6         4   \n",
      "43          0.000000                 5         3   \n",
      "44          0.416667                 7         5   \n",
      "\n",
      "    preferential_attachment_cluster  actual outcome  \n",
      "0                          0.695652               0  \n",
      "1                          1.000000               0  \n",
      "2                          0.173913               0  \n",
      "3                          0.000000               1  \n",
      "4                          0.130435               1  \n",
      "5                          0.695652               0  \n",
      "6                          0.173913               0  \n",
      "7                          0.565217               0  \n",
      "8                          0.521739               0  \n",
      "9                          0.347826               0  \n",
      "10                         0.000000               1  \n",
      "11                         1.000000               0  \n",
      "12                         0.695652               1  \n",
      "13                         0.347826               0  \n",
      "14                         1.000000               1  \n",
      "15                         0.782609               1  \n",
      "16                         0.695652               0  \n",
      "17                         1.000000               0  \n",
      "18                         0.347826               0  \n",
      "19                         1.000000               1  \n",
      "20                         0.391304               0  \n",
      "21                         1.000000               0  \n",
      "22                         0.695652               1  \n",
      "23                         0.173913               1  \n",
      "24                         0.173913               1  \n",
      "25                         0.695652               0  \n",
      "26                         0.521739               0  \n",
      "27                         0.782609               0  \n",
      "28                         0.173913               0  \n",
      "29                         0.173913               0  \n",
      "30                         0.695652               0  \n",
      "31                         0.521739               1  \n",
      "32                         1.000000               0  \n",
      "33                         0.695652               0  \n",
      "34                         1.000000               1  \n",
      "35                         0.347826               1  \n",
      "36                         0.391304               0  \n",
      "37                         0.521739               0  \n",
      "38                         0.695652               0  \n",
      "39                         0.782609               0  \n",
      "40                         0.521739               0  \n",
      "41                         0.695652               0  \n",
      "42                         0.521739               1  \n",
      "43                         0.130435               1  \n",
      "44                         1.000000               0  \n",
      "         Link Prediction Algorithm  Precision      AUPR\n",
      "0              num_common_neighbor   0.400000  0.495655\n",
      "1              jaccard_coefficient   0.266667  0.491998\n",
      "2              preferential_attach   0.400000  0.414908\n",
      "3               common_neighbor_tc   0.266667  0.283037\n",
      "4                 jaccard_coeff_tc   0.333333  0.267140\n",
      "5  preferential_attachment_cluster   0.266667  0.364363\n",
      "List of test edges: {('d10', 't1'), ('d11', 't7'), ('d6', 't4'), ('d9', 't3'), ('d14', 't5'), ('d15', 't6'), ('d11', 't4'), ('d6', 't5'), ('d3', 't3'), ('d8', 't7'), ('d7', 't3'), ('d13', 't7'), ('d6', 't2'), ('d9', 't6'), ('d4', 't1')}\n",
      "   d_node t_node  num_common_neighbor  jaccard_coefficient  \\\n",
      "0     d10     t1             0.444444             1.000000   \n",
      "1     d12     t5             0.444444             1.000000   \n",
      "2     d14     t4             0.222222             0.550000   \n",
      "3      d6     t4             0.444444             1.000000   \n",
      "4     d15     t5             0.222222             1.000000   \n",
      "5      d7     t3             1.000000             1.000000   \n",
      "6      d2     t5             0.444444             1.000000   \n",
      "7      d4     t5             0.333333             1.000000   \n",
      "8     d11     t2             0.555556             1.000000   \n",
      "9      d8     t5             0.333333             1.000000   \n",
      "10    d15     t4             0.444444             1.000000   \n",
      "11     d5     t2             0.666667             1.000000   \n",
      "12     d9     t3             0.777778             1.000000   \n",
      "13    d15     t6             0.777778             1.000000   \n",
      "14     d6     t1             0.333333             1.000000   \n",
      "15    d11     t4             0.444444             1.000000   \n",
      "16    d11     t6             0.666667             0.678571   \n",
      "17    d14     t1             0.000000             0.000000   \n",
      "18     d9     t2             0.666667             1.000000   \n",
      "19     d3     t3             0.777778             1.000000   \n",
      "20    d14     t3             0.333333             0.250000   \n",
      "21     d5     t4             0.555556             1.000000   \n",
      "22     d9     t6             0.888889             1.000000   \n",
      "23    d12     t1             0.555556             1.000000   \n",
      "24    d11     t7             0.444444             0.625000   \n",
      "25     d8     t6             0.777778             0.700000   \n",
      "26     d3     t4             0.555556             1.000000   \n",
      "27    d15     t1             0.333333             1.000000   \n",
      "28     d4     t7             0.666667             1.000000   \n",
      "29     d2     t7             0.777778             1.000000   \n",
      "30    d10     t2             0.666667             1.000000   \n",
      "31    d13     t5             0.333333             1.000000   \n",
      "32    d13     t3             0.777778             1.000000   \n",
      "33     d8     t7             0.555556             0.653846   \n",
      "34     d1     t5             0.444444             1.000000   \n",
      "35     d5     t1             0.444444             1.000000   \n",
      "36     d4     t1             0.444444             1.000000   \n",
      "37     d1     t4             0.666667             1.000000   \n",
      "38    d14     t5             0.000000             0.437500   \n",
      "39     d3     t1             0.444444             1.000000   \n",
      "40    d10     t7             0.666667             1.000000   \n",
      "41     d6     t5             0.222222             1.000000   \n",
      "42    d14     t2             0.333333             0.590909   \n",
      "43    d13     t7             0.666667             1.000000   \n",
      "44     d6     t2             0.555556             1.000000   \n",
      "\n",
      "    preferential_attach  common_neighbor_tc  number of d in cluster  \\\n",
      "0              0.333333                0.75                       5   \n",
      "1              0.375000                0.75                       5   \n",
      "2              0.083333                0.25                       3   \n",
      "3              0.250000                0.00                       2   \n",
      "4              0.125000                0.00                       2   \n",
      "5              1.000000                0.75                       5   \n",
      "6              0.375000                0.75                       5   \n",
      "7              0.250000                0.00                       2   \n",
      "8              0.312500                0.00                       2   \n",
      "9              0.250000                0.25                       3   \n",
      "10             0.250000                0.25                       2   \n",
      "11             0.500000                0.25                       3   \n",
      "12             0.583333                0.75                       5   \n",
      "13             0.437500                0.25                       2   \n",
      "14             0.187500                0.25                       2   \n",
      "15             0.250000                0.00                       2   \n",
      "16             0.437500                0.25                       2   \n",
      "17             0.041667                0.25                       3   \n",
      "18             0.500000                0.75                       5   \n",
      "19             0.583333                0.50                       3   \n",
      "20             0.166667                0.25                       3   \n",
      "21             0.416667                0.25                       3   \n",
      "22             0.666667                1.00                       5   \n",
      "23             0.479167                0.75                       5   \n",
      "24             0.312500                0.25                       2   \n",
      "25             0.666667                0.50                       3   \n",
      "26             0.416667                0.25                       3   \n",
      "27             0.187500                0.00                       2   \n",
      "28             0.500000                0.25                       2   \n",
      "29             0.687500                0.75                       5   \n",
      "30             0.500000                0.75                       5   \n",
      "31             0.250000                0.25                       3   \n",
      "32             0.583333                0.25                       3   \n",
      "33             0.500000                0.50                       3   \n",
      "34             0.375000                0.25                       3   \n",
      "35             0.333333                0.25                       3   \n",
      "36             0.333333                0.00                       2   \n",
      "37             0.583333                0.25                       3   \n",
      "38             0.000000                0.25                       3   \n",
      "39             0.333333                0.50                       3   \n",
      "40             0.500000                0.75                       5   \n",
      "41             0.125000                0.25                       2   \n",
      "42             0.125000                0.25                       3   \n",
      "43             0.500000                0.50                       3   \n",
      "44             0.312500                0.00                       2   \n",
      "\n",
      "    jaccard_coeff_tc  degree_d_cluster  degree_t  \\\n",
      "0           0.750000                 7         4   \n",
      "1           0.750000                 7         4   \n",
      "2           0.416667                 7         4   \n",
      "3           0.000000                 5         4   \n",
      "4           0.000000                 5         4   \n",
      "5           0.750000                 7         5   \n",
      "6           0.750000                 7         4   \n",
      "7           0.000000                 5         4   \n",
      "8           0.000000                 5         4   \n",
      "9           0.416667                 7         4   \n",
      "10          0.625000                 5         4   \n",
      "11          0.416667                 7         4   \n",
      "12          0.750000                 7         5   \n",
      "13          0.625000                 5         5   \n",
      "14          0.625000                 5         4   \n",
      "15          0.000000                 5         4   \n",
      "16          0.625000                 5         5   \n",
      "17          0.416667                 7         4   \n",
      "18          0.750000                 7         4   \n",
      "19          0.833333                 7         5   \n",
      "20          0.416667                 7         5   \n",
      "21          0.416667                 7         4   \n",
      "22          1.000000                 7         5   \n",
      "23          0.750000                 7         4   \n",
      "24          0.625000                 5         5   \n",
      "25          0.833333                 7         5   \n",
      "26          0.416667                 7         4   \n",
      "27          0.000000                 5         4   \n",
      "28          0.625000                 5         5   \n",
      "29          0.750000                 7         5   \n",
      "30          0.750000                 7         4   \n",
      "31          0.416667                 7         4   \n",
      "32          0.416667                 7         5   \n",
      "33          0.833333                 7         5   \n",
      "34          0.416667                 7         4   \n",
      "35          0.416667                 7         4   \n",
      "36          0.000000                 5         4   \n",
      "37          0.416667                 7         4   \n",
      "38          0.416667                 7         4   \n",
      "39          0.833333                 7         4   \n",
      "40          0.750000                 7         5   \n",
      "41          0.625000                 5         4   \n",
      "42          0.416667                 7         4   \n",
      "43          0.833333                 7         5   \n",
      "44          0.000000                 5         4   \n",
      "\n",
      "    preferential_attachment_cluster  actual outcome  \n",
      "0                          0.533333               1  \n",
      "1                          0.533333               0  \n",
      "2                          0.533333               0  \n",
      "3                          0.000000               1  \n",
      "4                          0.000000               0  \n",
      "5                          1.000000               1  \n",
      "6                          0.533333               0  \n",
      "7                          0.000000               0  \n",
      "8                          0.000000               0  \n",
      "9                          0.533333               0  \n",
      "10                         0.000000               0  \n",
      "11                         0.533333               0  \n",
      "12                         1.000000               1  \n",
      "13                         0.333333               1  \n",
      "14                         0.000000               0  \n",
      "15                         0.000000               1  \n",
      "16                         0.333333               0  \n",
      "17                         0.533333               0  \n",
      "18                         0.533333               0  \n",
      "19                         1.000000               1  \n",
      "20                         1.000000               0  \n",
      "21                         0.533333               0  \n",
      "22                         1.000000               1  \n",
      "23                         0.533333               0  \n",
      "24                         0.333333               1  \n",
      "25                         1.000000               0  \n",
      "26                         0.533333               0  \n",
      "27                         0.000000               0  \n",
      "28                         0.333333               0  \n",
      "29                         1.000000               0  \n",
      "30                         0.533333               0  \n",
      "31                         0.533333               0  \n",
      "32                         1.000000               0  \n",
      "33                         1.000000               1  \n",
      "34                         0.533333               0  \n",
      "35                         0.533333               0  \n",
      "36                         0.000000               1  \n",
      "37                         0.533333               0  \n",
      "38                         0.533333               1  \n",
      "39                         0.533333               0  \n",
      "40                         1.000000               0  \n",
      "41                         0.000000               1  \n",
      "42                         0.533333               0  \n",
      "43                         1.000000               1  \n",
      "44                         0.000000               1  \n",
      "         Link Prediction Algorithm  Precision      AUPR\n",
      "0              num_common_neighbor   0.400000  0.548322\n",
      "1              jaccard_coefficient   0.133333  0.600488\n",
      "2              preferential_attach   0.400000  0.433766\n",
      "3               common_neighbor_tc   0.400000  0.465201\n",
      "4                 jaccard_coeff_tc   0.466667  0.539465\n",
      "5  preferential_attachment_cluster   0.400000  0.505193\n",
      "List of test edges: {('d5', 't3'), ('d10', 't6'), ('d1', 't2'), ('d1', 't3'), ('d5', 't7'), ('d8', 't1'), ('d12', 't3'), ('d12', 't4'), ('d15', 't2'), ('d6', 't7'), ('d10', 't5'), ('d14', 't7'), ('d5', 't5'), ('d3', 't7'), ('d13', 't4')}\n",
      "   d_node t_node  num_common_neighbor  jaccard_coefficient  \\\n",
      "0     d12     t5                  0.3             1.000000   \n",
      "1     d14     t4                  0.2             0.400000   \n",
      "2     d12     t3                  0.6             0.769231   \n",
      "3     d15     t5                  0.3             1.000000   \n",
      "4     d15     t2                  0.5             1.000000   \n",
      "5     d12     t4                  0.4             0.727273   \n",
      "6      d6     t7                  0.7             1.000000   \n",
      "7      d2     t5                  0.5             1.000000   \n",
      "8     d14     t7                  0.3             0.700000   \n",
      "9      d5     t5                  0.0             0.571429   \n",
      "10     d4     t5                  0.5             1.000000   \n",
      "11    d11     t2                  0.7             1.000000   \n",
      "12     d5     t3                  0.2             0.181818   \n",
      "13     d8     t5                  0.3             0.700000   \n",
      "14    d15     t4                  0.5             1.000000   \n",
      "15     d5     t2                  0.2             0.666667   \n",
      "16     d6     t1                  0.7             1.000000   \n",
      "17    d11     t6                  1.0             0.823529   \n",
      "18    d14     t1                  0.3             0.700000   \n",
      "19     d9     t2                  0.8             1.000000   \n",
      "20    d14     t3                  0.4             0.500000   \n",
      "21     d5     t4                  0.0             0.000000   \n",
      "22    d10     t5                  0.2             0.666667   \n",
      "23    d12     t1                  0.4             0.727273   \n",
      "24     d8     t6                  0.8             0.625000   \n",
      "25     d3     t4                  0.6             1.000000   \n",
      "26     d5     t7                  0.1             0.333333   \n",
      "27    d15     t1                  0.5             1.000000   \n",
      "28     d4     t7                  0.7             1.000000   \n",
      "29     d2     t7                  0.7             1.000000   \n",
      "30    d10     t2                  0.4             0.727273   \n",
      "31    d13     t5                  0.4             1.000000   \n",
      "32    d13     t3                  0.8             1.000000   \n",
      "33     d1     t5                  0.3             1.000000   \n",
      "34     d5     t1                  0.1             0.333333   \n",
      "35    d10     t6                  0.6             0.400000   \n",
      "36     d3     t7                  0.6             1.000000   \n",
      "37    d13     t4                  0.6             1.000000   \n",
      "38     d1     t3                  0.7             1.000000   \n",
      "39     d1     t2                  0.5             1.000000   \n",
      "40     d8     t1                  0.6             1.000000   \n",
      "41     d1     t4                  0.5             1.000000   \n",
      "42     d3     t1                  0.6             1.000000   \n",
      "43    d10     t7                  0.4             0.727273   \n",
      "44    d14     t2                  0.3             0.700000   \n",
      "\n",
      "    preferential_attach  common_neighbor_tc  number of d in cluster  \\\n",
      "0              0.222222                0.50                       5   \n",
      "1              0.185185                0.00                       3   \n",
      "2              0.444444                1.00                       5   \n",
      "3              0.222222                0.00                       2   \n",
      "4              0.333333                0.25                       2   \n",
      "5              0.333333                1.00                       5   \n",
      "6              0.629630                0.25                       2   \n",
      "7              0.444444                0.50                       5   \n",
      "8              0.185185                0.25                       3   \n",
      "9              0.000000                0.25                       3   \n",
      "10             0.444444                0.00                       2   \n",
      "11             0.629630                0.25                       2   \n",
      "12             0.074074                0.00                       3   \n",
      "13             0.333333                0.25                       3   \n",
      "14             0.333333                0.25                       2   \n",
      "15             0.037037                0.25                       3   \n",
      "16             0.629630                0.25                       2   \n",
      "17             1.000000                0.25                       2   \n",
      "18             0.185185                0.25                       3   \n",
      "19             0.777778                0.75                       5   \n",
      "20             0.259259                0.00                       3   \n",
      "21             0.037037                0.00                       3   \n",
      "22             0.222222                0.50                       5   \n",
      "23             0.333333                1.00                       5   \n",
      "24             0.777778                0.50                       3   \n",
      "25             0.481481                0.25                       3   \n",
      "26             0.037037                0.25                       3   \n",
      "27             0.333333                0.25                       2   \n",
      "28             0.629630                0.25                       2   \n",
      "29             0.629630                0.75                       5   \n",
      "30             0.333333                0.75                       5   \n",
      "31             0.333333                0.25                       3   \n",
      "32             0.629630                0.00                       3   \n",
      "33             0.222222                0.25                       3   \n",
      "34             0.037037                0.25                       3   \n",
      "35             0.555556                1.00                       5   \n",
      "36             0.481481                0.50                       3   \n",
      "37             0.481481                0.00                       3   \n",
      "38             0.444444                0.50                       3   \n",
      "39             0.333333                0.50                       3   \n",
      "40             0.481481                0.25                       3   \n",
      "41             0.333333                0.25                       3   \n",
      "42             0.481481                0.25                       3   \n",
      "43             0.333333                0.75                       5   \n",
      "44             0.185185                0.25                       3   \n",
      "\n",
      "    jaccard_coeff_tc  degree_d_cluster  degree_t  \\\n",
      "0           0.500000                 7         4   \n",
      "1           0.000000                 5         4   \n",
      "2           1.000000                 7         4   \n",
      "3           0.000000                 6         4   \n",
      "4           0.625000                 6         5   \n",
      "5           1.000000                 7         4   \n",
      "6           0.625000                 7         5   \n",
      "7           0.500000                 7         4   \n",
      "8           0.416667                 5         5   \n",
      "9           0.416667                 5         4   \n",
      "10          0.000000                 6         4   \n",
      "11          0.625000                 7         5   \n",
      "12          0.000000                 5         4   \n",
      "13          0.416667                 7         4   \n",
      "14          0.625000                 6         4   \n",
      "15          0.416667                 5         5   \n",
      "16          0.625000                 7         5   \n",
      "17          0.625000                 7         5   \n",
      "18          0.416667                 5         5   \n",
      "19          0.750000                 7         5   \n",
      "20          0.000000                 5         4   \n",
      "21          0.000000                 5         4   \n",
      "22          0.500000                 7         4   \n",
      "23          1.000000                 7         5   \n",
      "24          0.833333                 7         5   \n",
      "25          0.416667                 7         4   \n",
      "26          0.416667                 5         5   \n",
      "27          0.625000                 6         5   \n",
      "28          0.625000                 6         5   \n",
      "29          0.750000                 7         5   \n",
      "30          0.750000                 7         5   \n",
      "31          0.416667                 5         4   \n",
      "32          0.000000                 5         4   \n",
      "33          0.416667                 7         4   \n",
      "34          0.416667                 5         5   \n",
      "35          1.000000                 7         5   \n",
      "36          0.833333                 7         5   \n",
      "37          0.000000                 5         4   \n",
      "38          0.833333                 7         4   \n",
      "39          0.833333                 7         5   \n",
      "40          0.416667                 7         5   \n",
      "41          0.416667                 7         4   \n",
      "42          0.416667                 7         5   \n",
      "43          0.750000                 7         5   \n",
      "44          0.416667                 5         5   \n",
      "\n",
      "    preferential_attachment_cluster  actual outcome  \n",
      "0                          0.533333               0  \n",
      "1                          0.000000               0  \n",
      "2                          0.533333               1  \n",
      "3                          0.266667               0  \n",
      "4                          0.666667               1  \n",
      "5                          0.533333               1  \n",
      "6                          1.000000               1  \n",
      "7                          0.533333               0  \n",
      "8                          0.333333               1  \n",
      "9                          0.000000               1  \n",
      "10                         0.266667               0  \n",
      "11                         1.000000               0  \n",
      "12                         0.000000               1  \n",
      "13                         0.533333               0  \n",
      "14                         0.266667               0  \n",
      "15                         0.333333               0  \n",
      "16                         1.000000               0  \n",
      "17                         1.000000               0  \n",
      "18                         0.333333               0  \n",
      "19                         1.000000               0  \n",
      "20                         0.000000               0  \n",
      "21                         0.000000               0  \n",
      "22                         0.533333               1  \n",
      "23                         1.000000               0  \n",
      "24                         1.000000               0  \n",
      "25                         0.533333               0  \n",
      "26                         0.333333               1  \n",
      "27                         0.666667               0  \n",
      "28                         0.666667               0  \n",
      "29                         1.000000               0  \n",
      "30                         1.000000               0  \n",
      "31                         0.000000               0  \n",
      "32                         0.000000               0  \n",
      "33                         0.533333               0  \n",
      "34                         0.333333               0  \n",
      "35                         1.000000               1  \n",
      "36                         1.000000               1  \n",
      "37                         0.000000               1  \n",
      "38                         0.533333               1  \n",
      "39                         1.000000               1  \n",
      "40                         1.000000               1  \n",
      "41                         0.533333               0  \n",
      "42                         1.000000               0  \n",
      "43                         1.000000               0  \n",
      "44                         0.333333               0  \n",
      "         Link Prediction Algorithm  Precision      AUPR\n",
      "0              num_common_neighbor   0.400000  0.296721\n",
      "1              jaccard_coefficient   0.333333  0.465928\n",
      "2              preferential_attach   0.333333  0.290574\n",
      "3               common_neighbor_tc   0.466667  0.496471\n",
      "4                 jaccard_coeff_tc   0.533333  0.556043\n",
      "5  preferential_attachment_cluster   0.333333  0.449613\n",
      "Average results:           Link Prediction Algorithm  Precision      AUPR\n",
      "0               common_neighbor_tc   0.413333  0.463847\n",
      "1                 jaccard_coeff_tc   0.413333  0.466849\n",
      "2              jaccard_coefficient   0.253333  0.565836\n",
      "3              num_common_neighbor   0.373333  0.405834\n",
      "4              preferential_attach   0.333333  0.358667\n",
      "5  preferential_attachment_cluster   0.346667  0.458117\n"
     ]
    }
   ],
   "source": [
    "# Evaluate Precision & AUPR for each cross validation by using the test set established earlier:\n",
    "# Iterate through each dataframe. For each dataframe, obtain rows with test set node pairs only (which should all be 1) normalize prediction scores, \n",
    "# Generate Precision & AUPR for that dataframe for each link prediction algorithm\n",
    "# Then obtain average AUC and AUPR for all dataframe to reach link prediction algorithm \n",
    "# References for understanding link prediction evaluation metrics:\n",
    "# Simone Daminelli et al 2015 New J. Phys. 17 113037 Main journal used for adopting precision & aupr over auc\n",
    "# Yang, Y., Lichtenwalter, R.N. & Chawla, N.V. Evaluating link prediction methods. Knowl Inf Syst 45, 751–782 (2015). States that ROC is deceptive. \n",
    "\n",
    "score_column = ['num_common_neighbor', \n",
    "                'jaccard_coefficient', \n",
    "                'preferential_attach', \n",
    "                'common_neighbor_tc', \n",
    "                'jaccard_coeff_tc', \n",
    "                'preferential_attachment_cluster']\n",
    "\n",
    "all_results = []\n",
    "for idx in range(5):\n",
    "    df_merge = df_dict_merged_dt[f'graph_{idx}'] # Get all missing edges from the training set graph\n",
    "    # Normalize each column\n",
    "    for column in score_column:\n",
    "        min_value = df_merge[column].min()\n",
    "        max_value = df_merge[column].max()\n",
    "        df_merge[column] = (df_merge[column] - min_value) / (max_value - min_value)\n",
    "\n",
    "    #print('df_merge_normalized: ', df_merge)\n",
    "    test_edges_set = set(dict_test_edges[f'test_edge_{idx}']) # Used to identify the test_edges\n",
    "    l = len(test_edges_set)    \n",
    "    print('List of test edges:', test_edges_set)\n",
    "    \n",
    "    # Create a new column in df_merge called 'actual outcome'\n",
    "    # For d node and t nodes that are within the list_test_edges, put 1 in the acutal outcome\n",
    "    # For the rest of the nodes, put 0\n",
    "    df_merge['actual outcome'] = df_merge.apply(lambda row: 1 if (row['d_node'], row['t_node']) in test_edges_set else 0, axis=1)\n",
    "    print(df_merge)\n",
    "\n",
    "\n",
    "    # Iterate through each prediction algorithm and calculate the metrics (precieion and aupr)\n",
    "    results = []\n",
    "    for score_type in score_column:\n",
    "        precision_at_L, aupr = calculate_metric(df_merge, score_type, l)\n",
    "        results.append({'Link Prediction Algorithm': score_type, 'Precision': precision_at_L, 'AUPR': aupr})\n",
    "        \n",
    "    # Convert the results to a DataFrame\n",
    "    results_df = pd.DataFrame(results)\n",
    "\n",
    "    # Append to all results\n",
    "    all_results.append(results_df)\n",
    "    \n",
    "    # Display the results\n",
    "    print(results_df)\n",
    "\n",
    "\n",
    "# Concatenate all results DataFrames into a single DataFrame\n",
    "concatenated_results = pd.concat(all_results)\n",
    "\n",
    "# Group by algorithm and calculate the mean of precision_at_L and aupr\n",
    "average_results = concatenated_results.groupby('Link Prediction Algorithm').mean().reset_index()\n",
    "    # Iterate through each prediction algorithm, by creating a dataframe where we have the d_node, t_node, prediction algorithm, and actual outcome\n",
    "    # sort rows based on the normalized score under the prediction algorithm\n",
    "    # Obtain precision value by analyzing how many 1s are there in the top L rows out of L rows\n",
    "    # Obtain AUPR with both column\n",
    "# Display the average results\n",
    "print('Average results: ', average_results)\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 30,
   "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>Link Prediction Algorithm</th>\n",
       "      <th>Precision</th>\n",
       "      <th>AUPR</th>\n",
       "    </tr>\n",
       "  </thead>\n",
       "  <tbody>\n",
       "    <tr>\n",
       "      <th>0</th>\n",
       "      <td>common_neighbor_tc</td>\n",
       "      <td>0.413333</td>\n",
       "      <td>0.463847</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>1</th>\n",
       "      <td>jaccard_coeff_tc</td>\n",
       "      <td>0.413333</td>\n",
       "      <td>0.466849</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>2</th>\n",
       "      <td>jaccard_coefficient</td>\n",
       "      <td>0.253333</td>\n",
       "      <td>0.565836</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>3</th>\n",
       "      <td>num_common_neighbor</td>\n",
       "      <td>0.373333</td>\n",
       "      <td>0.405834</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>4</th>\n",
       "      <td>preferential_attach</td>\n",
       "      <td>0.333333</td>\n",
       "      <td>0.358667</td>\n",
       "    </tr>\n",
       "    <tr>\n",
       "      <th>5</th>\n",
       "      <td>preferential_attachment_cluster</td>\n",
       "      <td>0.346667</td>\n",
       "      <td>0.458117</td>\n",
       "    </tr>\n",
       "  </tbody>\n",
       "</table>\n",
       "</div>"
      ],
      "text/plain": [
       "         Link Prediction Algorithm  Precision      AUPR\n",
       "0               common_neighbor_tc   0.413333  0.463847\n",
       "1                 jaccard_coeff_tc   0.413333  0.466849\n",
       "2              jaccard_coefficient   0.253333  0.565836\n",
       "3              num_common_neighbor   0.373333  0.405834\n",
       "4              preferential_attach   0.333333  0.358667\n",
       "5  preferential_attachment_cluster   0.346667  0.458117"
      ]
     },
     "execution_count": 30,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "average_results"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "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": "3.11.9"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 4
}