下面代码是使用python的分词和词云绘制的水浒top20人物出现的次数

from wordcloud import WordCloud

import jieba

import imageio

mask = imageio.imread('./su/1.png') #要绘制词云的形状

# 读取小说内容

with open('./su/水浒.txt', 'r', encoding='utf-8') as f:

words = f.read()

# counts用于存放出现的名字与次数

counts = {} # {‘曹操’:234，‘回寨’:56}

# 将不是人名(或者代表某个确定人，例孔明曰)的词放入excluds中

excludes = {"两个", "一个", "只见", "如何", "那里", "哥哥", "不能", "如此", "说道",

"军马", "次日", "军士", "头领", "左右", "一齐", "引兵", "里面", "梁山泊",

"只得", "于是", "今日", "不敢", "知府", "都督", "人马", "不知", "先锋",

"前面", "将来", "弟兄", "众人", "这里", "小人", "出来", "妇人", "好汉",

"便是", "问道", "起来", "却是", "我们", "因此", "甚么", "这个", "正是",

"三个", "且慢", "兄弟", "不是", "只是", "不曾", "且说", "不得", "一面",

"看时", "如今", "来到", "当下", "原来", "将军", "山寨", "喝道", "兄长",

"大喜", "天子", "商议", "那个", "商议", "公人", "性命", "城中", "喽罗",

"那厮", "小弟", "商议", "东京", "下山", "不见", "怎地", "上山", "随即",

"不要", "一条", "和尚", "背后", "洒家", "许多", "太尉", "答道", "收拾",

"却说", "一声"

}

# 分词

words_list = jieba.lcut(words)

# print(words_list)

# 将分词后的词语和创新次数保存在counts中

for word in words_list:

# 如果长度小于1代表不是人名，就跳出本次循环

if len(word) <= 1:

continue

else:

# 更新字典中的值

# counts[word] = 取出字典中原来键对应的值 + 1

# counts[word] = counts[word] + 1 # counts[word]如果没有就要报错

# 字典c.get(k) 如果字典中没有这个键 返回 NONE

counts[word] = counts.get(word, 0) + 1

# 将不是人名的词语删除

for word in excludes:

del counts[word]

# 排序

items = list(counts.items())

def sort_by_count(x):

return x[1]

items.sort(key=sort_by_count, reverse=True)

print(items)

# 将top20取出来

li = []

for i in range(20):

# 序列解包

role, count = items[i] #role代表人名，count是次数

print(role, count)

for _ in range(count): # _是告诉看代码的人循环里不需要使用临时变量

li.append(role)

# 将前十的人名以图片的形式保存到top.png中

text = ' '.join(li) # 将列表转换为字符串

WordCloud(

font_path='msyh.ttc', # 如果不加这句代码中文显示不出来

background_color='white',

width=800,

height=600,

mask=mask,

# 相邻两个值的重复

collocations=False

).generate(text).to_file('水浒Top20.png')

效果图

水浒Top20.png