人物关系提取

import os,sys

import jieba, codecs, math

import jieba.posseg as pseg



names = {} #姓名字典

relationships = {} #关系字典

lineNames = [] #每段内人物关系



jieba.load_userdict("dict.txt") #加载字典

with codecs.open("busan.txt","r","utf8") as f:

    for line in f.readlines():#读取文件所有东西，每行作为列表的一个元素

        poss = pseg.cut(line)#分词并返回该词词性

        lineNames.append([])#为新读入的一段添加人物名称列表


        for w in poss:

            if w.flag != "nr" or len(w.word)<2:

                continue

            #当分词长度小于2或该词词性不为nr时认为该词不为人名

            lineNames[-1].append(w.word)

            #为当段的环境增加一个人物

            if names.get(w.word) is None:

                names[w.word] = 0

                relationships[w.word] = {}


            names[w.word]+=1

            #该人物出现次数加1


for name, times in names.items():

    print(name, times)


print(relationships)

print(lineNames)

for line in lineNames:

    for name1 in line:

        for name2 in line:

            if name1 == name2:

                continue

            if relationships[name1].get(name2) is None:

                relationships[name1][name2]= 1

            else:

                relationships[name1][name2] = relationships[name1][name2]+1



print(relationships)

with codecs.open("busan_node.txt", "w", "gbk") as f:

    f.write("Id Label Weight\r\n")

    for name, times in names.items():

        f.write(name + " " +name+" "+str(times)+"\r\n")



with codecs.open("busan_edge.txt","w","gbk") as f:

    f.write("Source Target Weight\r\n")

    for name, edges in relationships.items():

        for v, w in edges.items():

            if w>3:

                f.write(name+" "+v+" "+str(w)