用 NetworkX 展示 Elasticsearch 同义词文件（qbit）

前言

• NetworkX github: https://github.com/networkx/networkx
• Elasticsearch solr 同义词格式： https://www.elastic.co/guide/en/elasticsearch/reference/7.17/...
• 技术栈

Python 3.11.8
NetworkX 3.3
matplotlib 3.9.2

同义词文件

• qbit_synonym.txt

怎=>怎,怎么,怎样
i-pod, i pod => ipod
事,事儿
事儿,事情
gpt,chatgpt

代码

• code

# encoding: utf8
# author: qbit
# date: 2024-08-20
# summary: 使用 NetworkX 展示 Elasticsearch 同义词文件

import matplotlib.pyplot as plt
import networkx as nx

plt.rcParams['font.sans-serif'] = ['SimHei']    # 避免中文乱码
G = nx.DiGraph()            # 创建一个有向图

with open('qbit_synonym.txt', mode='r', encoding='utf8') as f:
    for line in f:
        line = line.strip()
        if not line:
            continue
        if '=>' in line:    # 单向
            idx = line.index('=>')
            part1 = line[ : idx]
            part2 = line[idx+2: ]            
        else:               # 双向
            part1 = part2 = line
        for u in part1.split(','):
                u = u.strip()
                if not u:
                    continue
                for v in part2.split(','):
                    v = v.strip()
                    if not v:
                        continue
                    G.add_edge(u, v)

for node in G.nodes():
    print(node, end=f" -> {node}")
    for item in nx.descendants(G, node):
        print(f",{item}", end='')
    print()

pos = nx.circular_layout(G)

nx.draw_networkx(G, pos, alpha=0.5)     # 绘制图形
plt.show()

• 可以切换自己需要的布局

# pos = nx.planar_layout(G)
# pos = nx.random_layout(G)
# pos = nx.shell_layout(G)
# pos = nx.spring_layout(G)
# pos = nx.spectral_layout(G)
# pos = nx.spiral_layout(G)

结果

• 控制台打印

怎 -> 怎,怎么,怎样
怎么 -> 怎么
怎样 -> 怎样
i-pod -> i-pod,ipod
ipod -> ipod
i pod -> i pod,ipod
事 -> 事,事儿,事情
事儿 -> 事儿,事,事情
事情 -> 事情,事,事儿
gpt -> gpt,chatgpt
chatgpt -> chatgpt,gpt

• 图片展示
  

本文出自 qbit snap