python分析《三国演义》，谁才是这部书的绝对主角（包含统计指定角色的方法）

前面分析统计了金庸名著《倚天屠龙记》中人物按照出现次数并排序

https://www.cnblogs.com/becks/p/11421214.html

然后使用pyecharts，统计B站某视频弹幕内容，并绘制成词云显示

https://www.cnblogs.com/becks/p/14743080.html

这次，就用分析统计下《三国演义》这部书里各角色出现的频率，并绘制成词云显示，看看谁是绝对的主角吧

首先，我们需要把这部书里出现的人物都枚举出来，毕竟只统计角色信息，不需要把非人物名也统计进来

角色 = {'刘备','诸葛亮','关羽','张飞','刘禅',"孙权",'赵云','司马懿','周瑜','曹操','袁绍','马超','魏延',
        '黄忠','姜维','马岱','庞德','孟获','刘表','董卓','孙策',
        '鲁肃','司马昭','夏侯渊','王平','刘璋','袁术','吕蒙','甘宁','邓艾','曹仁',
        '陆逊','许褚','庞统','曹洪','李典','曹丕','廖化','曹真','吕布'}

然后就是读取实现准备好的《三国演义》书籍txt文档格式，使用jieba库对文档内容进行处理

# -*-coding:utf8-*-
# encoding:utf-8

import jieba   #倒入jieba库
import os
import sys
from collections import Counter#分词后词频统计
from pyecharts.charts import WordCloud#词云

path = os.path.abspath(os.path.dirname(sys.argv[0]))
txt=open(path+'\\171182.txt',"r", encoding='utf-8').read()    #读取三国演义文本

words=jieba.lcut(txt)   #jieba库分析文本
counts={}

在就是统计指定角色姓名出现次数

for word in words:
    if len(word)<=1:
        continue
    elif word in 角色:
        counts[word]=counts.get(word,0)+1
    else:
        None

绘制词云

items=list(counts.items())#字典到列表
wordcloud = WordCloud()
wordcloud.add("",items,word_size_range=[15, 80],rotate_step=30,shape='cardioid')
wordcloud.render(path+'\\wordcloud.html')

执行脚本后查看生成文件

曹操两个字的显示的最大，说明整部书里出现的次数最多。这肯定不对，罗贯中是刘备粉啊，

后来想了下，在三国里，直呼人姓名那是骂人，是损。那些所谓的正派人士都是有雅称的，比如卧龙、诸葛等等

改了下代码，把这些人的雅称也匹配进去

刘备 = {"玄德","玄德曰","先主","刘豫州","刘皇叔",'刘玄德','刘使君'}
诸葛亮 = {"孔明","孔明曰","卧龙","卧龙先生","诸葛先生",'孔明先生','诸葛丞相','诸葛'}
关羽 = {"关公","云长","汉寿亭侯","关云长"}
曹操 = {"孟德",'曹孟德','曹操'}
张飞 = {"张翼德",'翼德'}

同时，统计部分也作了处理

for word in words:    #筛选分析后的名词
    if len(word)<=1:   #因为词组中的汉字数大于1个即认为是一个词组，所以通过continue结束掉读取的汉字书为1的内容
        continue
    #elif word in exculdes:
        #continue
    #elif word in 诸葛亮 or word in 刘备 or word in 关羽 or word in 曹操:
        #counts[word]=counts.get(word,0)+1
    elif word in 刘备:
        word ="刘备"
        counts[word]=counts.get(word,0)+1
    elif word in 诸葛亮:
        word ="诸葛亮"
        counts[word]=counts.get(word,0)+1
    elif word in 曹操:
        word ="曹操"
        counts[word]=counts.get(word,0)+1
    elif word in 关羽:
        word ="关羽"
        counts[word]=counts.get(word,0)+1
    elif word in 张飞:
        word ="张飞"
        counts[word]=counts.get(word,0)+1
    elif word in 其他:
        counts[word]=counts.get(word,0)+1

    else:
        None

再次执行，嗯，诸葛亮是王者，诸葛亮合计出现了1350次，刘备合计出现1271次

附整个代码

# -*-coding:utf8-*-
# encoding:utf-8

import jieba   #倒入jieba库
import os
import sys
from collections import Counter#分词后词频统计
from pyecharts.charts import WordCloud#词云

path = os.path.abspath(os.path.dirname(sys.argv[0]))
txt=open(path+'\\三国演义.txt',"r", encoding='utf-8').read()    #文本

words=jieba.lcut(txt)   #jieba库分析文本
counts={}

刘备 = {"玄德","玄德曰","先主","刘豫州","刘皇叔",'刘玄德','刘使君'}
诸葛亮 = {"孔明","孔明曰","卧龙","卧龙先生","诸葛先生",'孔明先生','诸葛丞相','诸葛'}
关羽 = {"关公","云长","汉寿亭侯","关云长"}
刘禅 = {"后主"}
曹操 = {"孟德",'曹孟德','曹操'}
张飞 = {"张翼德",'翼德'}

其他 = {"孙权",'赵云','司马懿','周瑜','刘禅','袁绍','马超','魏延','黄忠','姜维','马岱','庞德','孟获','刘表','董卓','孙策',
        '鲁肃','司马昭','夏侯渊','王平','刘璋','袁术','吕蒙','甘宁','邓艾','曹仁','陆逊','许褚','庞统','曹洪','李典','曹丕','廖化','曹真','吕布'}
for word in words:    #筛选分析后的名词
    if len(word)<=1:   #因为词组中的汉字数大于1个即认为是一个词组，所以通过continue结束掉读取的汉字书为1的内容
        continue
    #elif word in exculdes:
        #continue
    #elif word in 诸葛亮 or word in 刘备 or word in 关羽 or word in 曹操:
        #counts[word]=counts.get(word,0)+1
    elif word in 刘备:
        word ="刘备"
        counts[word]=counts.get(word,0)+1
    elif word in 诸葛亮:
        word ="诸葛亮"
        counts[word]=counts.get(word,0)+1
    elif word in 曹操:
        word ="曹操"
        counts[word]=counts.get(word,0)+1
    elif word in 关羽:
        word ="关羽"
        counts[word]=counts.get(word,0)+1
    elif word in 张飞:
        word ="张飞"
        counts[word]=counts.get(word,0)+1
    elif word in 其他:
        counts[word]=counts.get(word,0)+1

    else:
        None

items=list(counts.items())#字典到列表

wordcloud = WordCloud()
wordcloud.add("",items,word_size_range=[15, 80],rotate_step=30,shape='cardioid')
wordcloud.render(path+'\\wordcloud.html')