基于Python的社交网络分析与图论算法实践

pip install networkx matplotlib

import networkx as nx
import matplotlib.pyplot as plt

# 创建一个空的无向图
G = nx.Graph()

# 添加节点
G.add_nodes_from(["Alice", "Bob", "Carol", "David"])

# 添加边
G.add_edges_from([("Alice", "Bob"), ("Bob", "Carol"), ("Carol", "David"), ("David", "Alice")])

# 绘制网络
pos = nx.spring_layout(G)  # 定义节点位置
nx.draw(G, pos, with_labels=True, node_color='skyblue', node_size=1500, edge_color='black', linewidths=1, font_size=15)

# 显示图形
plt.title("简单社交网络示例")
plt.show()

# 计算任意两点之间的最短路径
shortest_path = nx.shortest_path(G, source="Alice", target="Carol")
print("最短路径:", shortest_path)

# 计算节点的度中心性
degree_centrality = nx.degree_centrality(G)
print("节点的度中心性:", degree_centrality)

# 绘制度中心性分布图
plt.bar(degree_centrality.keys(), degree_centrality.values())
plt.title("节点的度中心性分布图")
plt.xlabel("节点")
plt.ylabel("度中心性")
plt.xticks(rotation=45)
plt.show()

# 计算节点的PageRank值
pagerank = nx.pagerank(G)
print("节点的PageRank值:", pagerank)

# 绘制PageRank值分布图
plt.bar(pagerank.keys(), pagerank.values())
plt.title("节点的PageRank值分布图")
plt.xlabel("节点")
plt.ylabel("PageRank值")
plt.xticks(rotation=45)
plt.show()

# 计算网络中的连通分量
connected_components = list(nx.connected_components(G))
print("网络中的连通分量:", connected_components)

# 绘制连通分量的可视化图
plt.figure(figsize=(8, 6))
colors = range(len(connected_components))
pos = nx.spring_layout(G)
for i, component in enumerate(connected_components):
    nx.draw_networkx_nodes(G, pos, nodelist=list(component), node_color=colors[i], node_size=500, alpha=0.8)
    nx.draw_networkx_edges(G, pos, alpha=0.5)
plt.title("网络中的连通分量")
plt.show()

# 使用Louvain算法进行社区发现
from community import community_louvain

partition = community_louvain.best_partition(G)
print("社区划分结果:", partition)

# 绘制社区划分结果的可视化图
plt.figure(figsize=(8, 6))
pos = nx.spring_layout(G)
cmap = plt.get_cmap('viridis', max(partition.values()) + 1)
nx.draw_networkx_nodes(G, pos, node_color=list(partition.values()), cmap=cmap, node_size=500, alpha=0.8)
nx.draw_networkx_edges(G, pos, alpha=0.5)
plt.title("社区划分结果")
plt.show()

import tweepy
from tweepy import OAuthHandler

# Twitter API的认证信息，请替换为你自己的
consumer_key = 'your_consumer_key'
consumer_secret = 'your_consumer_secret'
access_token = 'your_access_token'
access_secret = 'your_access_secret'

auth = OAuthHandler(consumer_key, consumer_secret)
auth.set_access_token(access_token, access_secret)
api = tweepy.API(auth)

# 获取某个用户的关注者列表
user = "realDonaldTrump"
followers = api.followers_ids(user)

# 创建一个空的有向图
G_twitter = nx.DiGraph()

# 添加用户和其关注者作为节点和边
for follower_id in followers:
    G_twitter.add_edge(user, str(follower_id))

# 绘制Twitter社交网络图
plt.figure(figsize=(10, 10))
pos = nx.spring_layout(G_twitter, k=0.1)
nx.draw(G_twitter, pos, with_labels=False, node_size=30, width=0.1, alpha=0.7)
plt.title("Twitter社交网络图")
plt.show()

# 示例：使用K-means聚类分析节点
from sklearn.cluster import KMeans
import numpy as np

# 提取节点的度中心性作为特征
centrality = np.array(list(nx.degree_centrality(G_twitter).values())).reshape(-1, 1)

# 使用K-means进行聚类分析
kmeans = KMeans(n_clusters=2, random_state=0).fit(centrality)

# 可视化聚类结果
plt.figure(figsize=(8, 6))
plt.scatter(range(len(centrality)), centrality, c=kmeans.labels_, cmap='viridis')
plt.title("节点度中心性的K-means聚类分析")
plt.xlabel("节点编号")
plt.ylabel("度中心性")
plt.show()