jieba+wordcloud 詞雲分析 202302 QCon 議題 TOP 關鍵詞

效果圖

步驟

（1）依賴

• python 庫

  pip install jieba wordcloud
             

• 資料
  • 概覽

  $ head -n 5 input.txt
  中國軟體技術發展洞察和趨勢預測報告 2023
  QCon 大會内容策劃思路
  FinOps：從概念到落地
  開源晶片的發展現狀、機遇和未來
  樂觀者前行，Infra 出海的挑戰與機遇
             

  • 來源：https://qcon.infoq.cn/202302/beijing/schedule
  • 提取方法：使用 Chrome 的 XPath 在控制台 Console 提取

    $x('//tbody/tr/td/a/text()').map(x => x.data)
               

  

（2）源碼

import jieba
import wordcloud


# 對文本進行分詞
# top_num 取前 top_num
# exclude_words 無關詞語清單
# user_words 自定義分詞
def cut_word(input_path, out_path, top_num=30, exclude_words=[], user_words=[]):
    file = open(input_path, 'r', encoding='utf-8')
    txt = file.read()
    if len(user_words) > 0:
        for user_word in user_words:
            jieba.add_word(user_word)
    words = jieba.lcut(txt)
    # 對詞頻進行統計
    count = {}
    for word in words:
        if len(word) == 1:
            continue
        else:
            count[word] = count.get(word, 0) + 1

    # 周遊字典的所有鍵，即所有 word
    for key in list(count.keys()):
        # 引入停用詞
        if key in exclude_words:
            del count[key]
    lists = list(count.items())
    # 詞頻排序
    lists.sort(key=lambda x: x[1], reverse=True)
    # 列印前 top_num 條詞頻
    for i in range(top_num):
        word, number = lists[i]
        print("關鍵字：{:-<5}頻次：{}".format(word, number))

    # 詞頻寫入
    with open(out_path, 'w', encoding='utf-8') as f:
        for i in range(top_num):
            word, number = lists[i]
            f.write('{}\t{}\n'.format(word, number))
        f.close()
        return out_path


# 制作詞雲
def get_cloud(input_path, image_out_path):
    with open(input_path, 'r', encoding='utf-8') as f:
        text = f.read()
    wcloud = wordcloud.WordCloud(
        font_path=r'‪C:\Windows\Fonts\simhei.ttf',
        background_color='white',
        width=500,
        max_words=1000,
        height=400,
        margin=5
    ).generate(text)

    # 指定詞雲檔案路徑
    wcloud.to_file(image_out_path)
    f.close()
    print("詞雲圖檔已儲存")


if __name__ == '__main__':
    cut_word('./input.txt', out_path='./wordcloud.txt', top_num=200, exclude_words=[
        '實踐', '技術', '基于', '應用', '建設', '實戰', '探索', '系統', '體系'
    ], user_words=['雲原生', '研發效能', '分布式', '微服務'])

    get_cloud(input_path='./wordcloud.txt', image_out_path='./qcon.png')
           

參考

• https://www.cnblogs.com/yangyezhuang/p/16896980.html
• https://blog.csdn.net/zhangzeyuaaa/article/details/122192065
• https://baijiahao.baidu.com/s?id=1702691581630693235