程序设计实训 - 题目 12 词频统计

[2019301248-侯星宇]

程序设计实训 - Python

题目 12

代码实现

# NPU 程序设计实训 HW12 词频统计
# 函数 word_freq() 用于统计词频
import string
import re


def word_freq(path):
    # 读取高频词
    sight_word = open('./sight word.txt', 'r').read()
    sight_word = sight_word.lower()
    sight_word_list = sight_word.split()
    # 读取文本
    text = open(path, 'r').read()
    text_list = re.findall(r'[a-z\'0-9—“”]+', text.lower())
    text_list = [i.lower() for i in text_list if i and i.lower() not in sight_word_list]
    dic = {}
    # 统计词频
    for i in text_list:
        if i in dic:
            dic[i] += 1
        else:
            dic[i] = 1
    result = sorted(dic.items(), key=lambda x: (x[1], x[0]), reverse=True)

    return result[:10]

结果展示

# 测试函数
path = './text.txt'
print(word_freq(path))

结果如下