要求及数据集

hw0.pdfWiki-Vote.txtstackoverflow-Java.txt
内容很简单

代码

Wiki

import snap
import matplotlib.pyplot as plt
def init_graph():
    G = snap.TNGraph.New()
    with open('Wiki-Vote.txt', 'r', encoding='utf-8') as f:
        for line in f.readlines():
            if line[0] == '#':
                continue
            node_in, node_out = line.split('\t')
            if not G.IsNode(int(node_in)):
                G.AddNode(int(node_in))
            if not G.IsNode(int(node_out)):
                G.AddNode(int(node_out))
            G.AddEdge(int(node_in), int(node_out))
    return G
def analyze_graph(G):
    print('nodes in network:')
    total = G.GetNodes()
    print(total)
    print('number of nodes with self edge:')
    num = 0
    for node in G.Nodes():
        if node.IsOutNId(node.GetId()):
            num += 1
    print(num)
    print('number of directed edges:', total - num)
    print('number of undirected edges:')
    num = 0
    for edge in G.Edges():
        node_1 = edge.GetSrcNId()
        node_2 = edge.GetDstNId()
        if not G.IsEdge(node_2, node_1):
            num += 1
    print(num)
    print('number of reciprocated edges:', G.GetEdges() - num)
    num_in1, num_out1 = 0, 0
    num_in2, num_out2 = 0, 0
    for node in G.Nodes():
        in_deg = node.GetInDeg()
        out_deg = node.GetOutDeg()
        if not in_deg:
            num_in1 += 1
        elif in_deg > 10:
            num_in2 += 1
        if not out_deg:
            num_out1 += 1
        elif out_deg > 10:
            num_out2 += 1
    print('zero in deg:', num_in1)
    print('zero out deg', num_out1)
    print(num_in2, num_out2)
def analyze_further(G):
    X, Y = [], []
    map = {}
    for node in G.Nodes():
        out_deg = node.GetOutDeg()
        if out_deg not in map:
            map[out_deg] = 1
        else:
            map[out_deg] += 1
    X = [item for item in map.keys()]
    X.sort()
    Y = [map[x] for x in X]
    plt.loglog(X, Y, linestyle='dotted', color='b', label='Wiki Network')
    plt.xlabel('Node Degree (log)')
    plt.ylabel('Proportion of Nodes with a Given Degree (log)')
    plt.title('Degree Distribution of Wiki-vote')
    plt.legend()
    plt.show()
if __name__ == '__main__':
    G = init_graph()
    analyze_graph(G)
    analyze_further(G)

Java

import snap
def init_graph():
    G = snap.TNGraph().New()
    with open('stackoverflow-Java.txt', 'r', encoding='utf-8') as f:
        for line in f.readlines():
            node_in, node_out = line.split('\t')
            if not G.IsNode(int(node_in)):
                G.AddNode(int(node_in))
            if not G.IsNode(int(node_out)):
                G.AddNode(int(node_out))
            G.AddEdge(int(node_in), int(node_out))
    return G
def analyze_graph(G):
    print('weakly connected components:', len(G.GetWccs()))
    print('The number of edges and the number of nodes in the largest weakly connected component:')
    component = G.GetMxWcc()
    print(component.GetNodes(), component.GetEdges())
    print('IDs of the top 3 most central nodes in the network by PagePank scores:')
    PRankH = G.GetPageRank()
    values = PRankH.values()
    values.sort()
    for item in PRankH:
        if PRankH[item] == values[-1]:
            print(item)
        if PRankH[item] == values[-2]:
            print(item)
        if PRankH[item] == values[-3]:
            print(item)
    print('IDs of the top 3 hubs and top 3 authorities in the network by HITS scores',)
    NIdHubH, NIdAuthH = G.GetHits()
    values = NIdHubH.values()
    values.sort()
    for item in NIdHubH:
        if NIdHubH[item] == values[-1]:
            print(item)
        if NIdHubH[item] == values[-2]:
            print(item)
        if NIdHubH[item] == values[-3]:
            print(item)
    values = NIdAuthH.values()
    values.sort()
    for item in NIdAuthH:
        if NIdAuthH[item] == values[-1]:
            print(item)
        if NIdAuthH[item] == values[-2]:
            print(item)
        if NIdAuthH[item] == values[-3]:
            print(item)
if __name__ == '__main__':
    G = init_graph()
    analyze_graph(G)