{"id":1117200,"date":"2025-01-08T18:23:48","date_gmt":"2025-01-08T10:23:48","guid":{"rendered":"https:\/\/docs.pingcode.com\/ask\/ask-ask\/1117200.html"},"modified":"2025-01-08T18:23:51","modified_gmt":"2025-01-08T10:23:51","slug":"python%e5%a6%82%e4%bd%95%e5%af%b9%e5%88%86%e8%af%8d%e5%90%8e%e7%9a%84%e6%96%87%e6%9c%ac%e8%81%9a%e7%b1%bb","status":"publish","type":"post","link":"https:\/\/docs.pingcode.com\/ask\/1117200.html","title":{"rendered":"python\u5982\u4f55\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u805a\u7c7b"},"content":{"rendered":"

\"python\u5982\u4f55\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u805a\u7c7b\"<\/p>\n

Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\u7684\u65b9\u6cd5\u6709\u5f88\u591a\u79cd\uff0c\u4e3b\u8981\u5305\u62ecTF-IDF\u5411\u91cf\u5316\u3001K-means\u805a\u7c7b\u3001DBSCAN\u805a\u7c7b\u3001\u5c42\u6b21\u805a\u7c7b\u7b49\u65b9\u6cd5\u3002<\/strong> \u5176\u4e2d\uff0cTF-IDF\u5411\u91cf\u5316<\/strong>\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u6587\u672c\u8868\u793a\u65b9\u6cd5\uff0c\u901a\u8fc7\u8ba1\u7b97\u6bcf\u4e2a\u8bcd\u7684\u91cd\u8981\u6027\u6765\u8868\u793a\u6587\u672c\uff1bK-means\u805a\u7c7b<\/strong>\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u65e0\u76d1\u7763\u673a\u5668\u5b66\u4e60<\/a>\u7b97\u6cd5\uff0c\u9002\u7528\u4e8e\u5927\u89c4\u6a21\u6570\u636e\u96c6\u7684\u805a\u7c7b\u4efb\u52a1\u3002\u4e0b\u9762\u5c06\u8be6\u7ec6\u4ecb\u7ecd\u4f7f\u7528Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\u7684\u5177\u4f53\u6b65\u9aa4\u548c\u65b9\u6cd5\u3002<\/p>\n<\/p>\n

\u4e00\u3001\u6587\u672c\u9884\u5904\u7406<\/p>\n<\/p>\n

\u5728\u8fdb\u884c\u805a\u7c7b\u4e4b\u524d\uff0c\u9996\u5148\u9700\u8981\u5bf9\u6587\u672c\u8fdb\u884c\u9884\u5904\u7406\u3002\u6587\u672c\u9884\u5904\u7406\u4e3b\u8981\u5305\u62ec\u5206\u8bcd\u3001\u53bb\u505c\u7528\u8bcd\u3001\u8bcd\u5e72\u5316\u6216\u8bcd\u5f62\u8fd8\u539f\u7b49\u6b65\u9aa4\u3002\u53ef\u4ee5\u4f7f\u7528Python\u7684NLTK\u3001spaCy\u6216jieba\u7b49\u5e93\u6765\u8fdb\u884c\u6587\u672c\u9884\u5904\u7406\u3002<\/p>\n<\/p>\n

    \n
  1. \u5206\u8bcd<\/li>\n<\/ol>\n

    \u5206\u8bcd\u662f\u5c06\u6587\u672c\u5207\u5206\u6210\u5355\u72ec\u7684\u8bcd\u8bed\u6216\u8bcd\u7ec4\u3002\u5bf9\u4e8e\u82f1\u6587\u6587\u672c\uff0c\u53ef\u4ee5\u4f7f\u7528NLTK\u6216spaCy\u5e93\uff1b\u5bf9\u4e8e\u4e2d\u6587\u6587\u672c\uff0c\u53ef\u4ee5\u4f7f\u7528jieba\u5e93\u3002<\/p>\n<\/p>\n

    import jieba<\/p>\n
    \u793a\u4f8b\u6587\u672c<\/strong><\/h2>\n
    text = "Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\u7684\u65b9\u6cd5\u6709\u54ea\u4e9b\uff1f"<\/p>\n
    \u4f7f\u7528jieba\u8fdb\u884c\u5206\u8bcd<\/strong><\/h2>\n
    words = jieba.lcut(text)<\/p>\n
    print(words)<\/p>\n
    <\/code><\/pre>\n<\/p>\n
      \n
    1. \u53bb\u505c\u7528\u8bcd<\/li>\n<\/ol>\n
      \u505c\u7528\u8bcd\u662f\u6307\u5728\u6587\u672c\u5904\u7406\u4e2d\u88ab\u8fc7\u6ee4\u6389\u7684\u5e38\u89c1\u8bcd\u8bed\uff0c\u5982\u201c\u7684\u201d\u3001\u201c\u4e86\u201d\u3001\u201c\u662f\u201d\u7b49\u3002\u53bb\u505c\u7528\u8bcd\u53ef\u4ee5\u51cf\u5c11\u566a\u97f3\uff0c\u63d0\u9ad8\u805a\u7c7b\u6548\u679c\u3002<\/p>\n<\/p>\n
      from nltk.corpus import stopwords<\/p>\n
      \u4e0b\u8f7d\u505c\u7528\u8bcd\u5217\u8868<\/strong><\/h2>\n
      import nltk<\/p>\n
      nltk.download('stopwords')<\/p>\n
      \u83b7\u53d6\u82f1\u6587\u505c\u7528\u8bcd\u5217\u8868<\/strong><\/h2>\n
      stop_words = set(stopwords.words('english'))<\/p>\n
      \u8fc7\u6ee4\u505c\u7528\u8bcd<\/strong><\/h2>\n
      filtered_words = [word for word in words if word not in stop_words]<\/p>\n
      print(filtered_words)<\/p>\n
      <\/code><\/pre>\n<\/p>\n
        \n
      1. \u8bcd\u5e72\u5316\u6216\u8bcd\u5f62\u8fd8\u539f<\/li>\n<\/ol>\n
        \u8bcd\u5e72\u5316\u662f\u5c06\u5355\u8bcd\u8fd8\u539f\u4e3a\u5176\u8bcd\u5e72\u5f62\u5f0f\uff0c\u8bcd\u5f62\u8fd8\u539f\u662f\u5c06\u5355\u8bcd\u8fd8\u539f\u4e3a\u5176\u57fa\u672c\u5f62\u5f0f\u3002\u53ef\u4ee5\u4f7f\u7528NLTK\u6216spaCy\u5e93\u8fdb\u884c\u8bcd\u5e72\u5316\u6216\u8bcd\u5f62\u8fd8\u539f\u3002<\/p>\n<\/p>\n
        from nltk.stem import PorterStemmer<\/p>\n
        \u521d\u59cb\u5316PorterStemmer<\/strong><\/h2>\n
        stemmer = PorterStemmer()<\/p>\n
        \u8fdb\u884c\u8bcd\u5e72\u5316<\/strong><\/h2>\n
        stemmed_words = [stemmer.stem(word) for word in filtered_words]<\/p>\n
        print(stemmed_words)<\/p>\n
        <\/code><\/pre>\n<\/p>\n
        \u4e8c\u3001\u6587\u672c\u8868\u793a<\/p>\n<\/p>\n
        \u6587\u672c\u8868\u793a\u662f\u5c06\u6587\u672c\u8f6c\u6362\u4e3a\u6570\u503c\u5411\u91cf\uff0c\u4ee5\u4fbf\u8fdb\u884c\u805a\u7c7b\u3002\u5e38\u7528\u7684\u6587\u672c\u8868\u793a\u65b9\u6cd5\u6709TF-IDF\u5411\u91cf\u5316\u548c\u8bcd\u5d4c\u5165\uff08\u5982Word2Vec\u3001GloVe\u7b49\uff09\u3002<\/p>\n<\/p>\n
          \n
        1. TF-IDF\u5411\u91cf\u5316<\/li>\n<\/ol>\n
          TF-IDF\uff08Term Frequency-Inverse Document Frequency\uff09\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u6587\u672c\u8868\u793a\u65b9\u6cd5\uff0c\u901a\u8fc7\u8ba1\u7b97\u8bcd\u8bed\u5728\u6587\u6863\u4e2d\u7684\u91cd\u8981\u6027\u6765\u8868\u793a\u6587\u672c\u3002<\/p>\n<\/p>\n
          from sklearn.feature_extraction.text import TfidfVectorizer<\/p>\n
          \u793a\u4f8b\u6587\u672c\u5217\u8868<\/strong><\/h2>\n
          documents = ["Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\u7684\u65b9\u6cd5\u6709\u54ea\u4e9b\uff1f", "\u4f7f\u7528TF-IDF\u8fdb\u884c\u6587\u672c\u8868\u793a", "K-means\u805a\u7c7b\u7b97\u6cd5"]<\/p>\n
          \u521d\u59cb\u5316TfidfVectorizer<\/strong><\/h2>\n
          vectorizer = TfidfVectorizer()<\/p>\n
          \u8fdb\u884cTF-IDF\u5411\u91cf\u5316<\/strong><\/h2>\n
          tfidf_matrix = vectorizer.fit_transform(documents)<\/p>\n
          print(tfidf_matrix.toarray())<\/p>\n
          <\/code><\/pre>\n<\/p>\n
            \n
          1. \u8bcd\u5d4c\u5165<\/li>\n<\/ol>\n
            \u8bcd\u5d4c\u5165\u662f\u5c06\u8bcd\u8bed\u8868\u793a\u4e3a\u4f4e\u7ef4\u5411\u91cf\uff0c\u53ef\u4ee5\u4f7f\u7528Word2Vec\u3001GloVe\u7b49\u9884\u8bad\u7ec3\u8bcd\u5d4c\u5165\u6a21\u578b\u3002<\/p>\n<\/p>\n
            from gensim.models import Word2Vec<\/p>\n
            \u793a\u4f8b\u5206\u8bcd\u540e\u7684\u6587\u672c\u5217\u8868<\/strong><\/h2>\n
            tokenized_documents = [["Python", "\u5bf9", "\u5206\u8bcd", "\u540e", "\u7684", "\u6587\u672c", "\u8fdb\u884c", "\u805a\u7c7b"], ["\u4f7f\u7528", "TF-IDF", "\u8fdb\u884c", "\u6587\u672c", "\u8868\u793a"], ["K-means", "\u805a\u7c7b", "\u7b97\u6cd5"]]<\/p>\n
            \u8bad\u7ec3Word2Vec\u6a21\u578b<\/strong><\/h2>\n
            model = Word2Vec(sentences=tokenized_documents, vector_size=100, window=5, min_count=1, workers=4)<\/p>\n
            \u83b7\u53d6\u8bcd\u8bed\u5411\u91cf<\/strong><\/h2>\n
            word_vector = model.wv['Python']<\/p>\n
            print(word_vector)<\/p>\n
            <\/code><\/pre>\n<\/p>\n
            \u4e09\u3001\u6587\u672c\u805a\u7c7b<\/p>\n<\/p>\n
            \u6587\u672c\u805a\u7c7b\u662f\u5c06\u76f8\u4f3c\u7684\u6587\u672c\u5206\u4e3a\u4e00\u7c7b\uff0c\u5e38\u7528\u7684\u805a\u7c7b\u7b97\u6cd5\u6709K-means\u805a\u7c7b\u3001DBSCAN\u805a\u7c7b\u3001\u5c42\u6b21\u805a\u7c7b\u7b49\u3002<\/p>\n<\/p>\n
              \n
            1. K-means\u805a\u7c7b<\/li>\n<\/ol>\n
              K-means\u805a\u7c7b\u662f\u4e00\u79cd\u5e38\u89c1\u7684\u65e0\u76d1\u7763\u673a\u5668\u5b66\u4e60\u7b97\u6cd5\uff0c\u9002\u7528\u4e8e\u5927\u89c4\u6a21\u6570\u636e\u96c6\u7684\u805a\u7c7b\u4efb\u52a1\u3002<\/p>\n<\/p>\n
              from sklearn.cluster import KMeans<\/p>\n
              \u521d\u59cb\u5316KMeans\u805a\u7c7b\u6a21\u578b<\/strong><\/h2>\n
              kmeans = KMeans(n_clusters=3, random_state=42)<\/p>\n
              \u8fdb\u884c\u805a\u7c7b<\/strong><\/h2>\n
              kmeans.fit(tfidf_matrix)<\/p>\n
              \u83b7\u53d6\u805a\u7c7b\u7ed3\u679c<\/strong><\/h2>\n
              labels = kmeans.labels_<\/p>\n
              print(labels)<\/p>\n
              <\/code><\/pre>\n<\/p>\n
                \n
              1. DBSCAN\u805a\u7c7b<\/li>\n<\/ol>\n
                DBSCAN\uff08Density-Based Spatial Clustering of Applications with Noise\uff09\u662f\u4e00\u79cd\u57fa\u4e8e\u5bc6\u5ea6\u7684\u805a\u7c7b\u7b97\u6cd5\uff0c\u9002\u7528\u4e8e\u5904\u7406\u5177\u6709\u566a\u97f3\u7684\u6570\u636e\u96c6\u3002<\/p>\n<\/p>\n
                from sklearn.cluster import DBSCAN<\/p>\n
                \u521d\u59cb\u5316DBSCAN\u805a\u7c7b\u6a21\u578b<\/strong><\/h2>\n
                dbscan = DBSCAN(eps=0.5, min_samples=5)<\/p>\n
                \u8fdb\u884c\u805a\u7c7b<\/strong><\/h2>\n
                dbscan.fit(tfidf_matrix)<\/p>\n
                \u83b7\u53d6\u805a\u7c7b\u7ed3\u679c<\/strong><\/h2>\n
                labels = dbscan.labels_<\/p>\n
                print(labels)<\/p>\n
                <\/code><\/pre>\n<\/p>\n
                  \n
                1. \u5c42\u6b21\u805a\u7c7b<\/li>\n<\/ol>\n
                  \u5c42\u6b21\u805a\u7c7b\u662f\u4e00\u79cd\u57fa\u4e8e\u6811\u72b6\u7ed3\u6784\u7684\u805a\u7c7b\u7b97\u6cd5\uff0c\u53ef\u4ee5\u751f\u6210\u5c42\u6b21\u5316\u7684\u805a\u7c7b\u7ed3\u679c\u3002<\/p>\n<\/p>\n
                  from scipy.cluster.hierarchy import linkage, dendrogram<\/p>\n
                  \u8fdb\u884c\u5c42\u6b21\u805a\u7c7b<\/strong><\/h2>\n
                  Z = linkage(tfidf_matrix.toarray(), method='ward')<\/p>\n
                  \u7ed8\u5236\u6811\u72b6\u56fe<\/strong><\/h2>\n
                  import matplotlib.pyplot as plt<\/p>\n
                  plt.figure()<\/p>\n
                  dendrogram(Z)<\/p>\n
                  plt.show()<\/p>\n
                  <\/code><\/pre>\n<\/p>\n
                  \u56db\u3001\u805a\u7c7b\u7ed3\u679c\u5206\u6790<\/p>\n<\/p>\n
                  \u805a\u7c7b\u7ed3\u679c\u5206\u6790\u662f\u5bf9\u805a\u7c7b\u7ed3\u679c\u8fdb\u884c\u89e3\u91ca\u548c\u8bc4\u4ef7\uff0c\u4ee5\u786e\u5b9a\u805a\u7c7b\u6548\u679c\u3002\u5e38\u7528\u7684\u65b9\u6cd5\u6709\u8f6e\u5ed3\u7cfb\u6570\u3001\u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027\u7b49\u3002<\/p>\n<\/p>\n
                    \n
                  1. \u8f6e\u5ed3\u7cfb\u6570<\/li>\n<\/ol>\n
                    \u8f6e\u5ed3\u7cfb\u6570\uff08Silhouette Coefficient\uff09\u662f\u4e00\u79cd\u8bc4\u4f30\u805a\u7c7b\u6548\u679c\u7684\u6307\u6807\uff0c\u503c\u8d8a\u5927\u8868\u793a\u805a\u7c7b\u6548\u679c\u8d8a\u597d\u3002<\/p>\n<\/p>\n
                    from sklearn.metrics import silhouette_score<\/p>\n
                    \u8ba1\u7b97\u8f6e\u5ed3\u7cfb\u6570<\/strong><\/h2>\n
                    silhouette_avg = silhouette_score(tfidf_matrix, labels)<\/p>\n
                    print("\u8f6e\u5ed3\u7cfb\u6570:", silhouette_avg)<\/p>\n
                    <\/code><\/pre>\n<\/p>\n
                      \n
                    1. \u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027<\/li>\n<\/ol>\n
                      \u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027\u662f\u8bc4\u4f30\u805a\u7c7b\u6548\u679c\u7684\u53e6\u4e00\u4e2a\u6307\u6807\uff0c\u8868\u793a\u540c\u4e00\u805a\u7c7b\u5185\u7684\u6570\u636e\u70b9\u4e4b\u95f4\u7684\u76f8\u4f3c\u5ea6\u3002<\/p>\n<\/p>\n
                      from sklearn.metrics import calinski_harabasz_score<\/p>\n
                      \u8ba1\u7b97\u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027<\/strong><\/h2>\n
                      ch_score = calinski_harabasz_score(tfidf_matrix.toarray(), labels)<\/p>\n
                      print("\u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027:", ch_score)<\/p>\n
                      <\/code><\/pre>\n<\/p>\n
                      \u4e94\u3001\u6848\u4f8b\u5b9e\u6218<\/p>\n<\/p>\n
                      \u4e3a\u4e86\u66f4\u597d\u5730\u7406\u89e3\u5982\u4f55\u4f7f\u7528Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\uff0c\u4e0b\u9762\u901a\u8fc7\u4e00\u4e2a\u5177\u4f53\u7684\u6848\u4f8b\u6765\u6f14\u793a\u6574\u4e2a\u8fc7\u7a0b\u3002<\/p>\n<\/p>\n
                      \u6848\u4f8b\u80cc\u666f\uff1a\u5047\u8bbe\u6211\u4eec\u6709\u4e00\u7ec4\u65b0\u95fb\u6587\u7ae0\uff0c\u76ee\u6807\u662f\u5c06\u8fd9\u4e9b\u6587\u7ae0\u6309\u7167\u4e3b\u9898\u8fdb\u884c\u805a\u7c7b\u3002<\/p>\n<\/p>\n
                        \n
                      1. \u6570\u636e\u51c6\u5907<\/li>\n<\/ol>\n
                        \u9996\u5148\u51c6\u5907\u4e00\u7ec4\u65b0\u95fb\u6587\u7ae0\u6570\u636e\u3002<\/p>\n<\/p>\n
                        news_articles = [<\/p>\n
                            "\u82f9\u679c\u516c\u53f8\u53d1\u5e03\u4e86\u65b0\u6b3eiPhone",<\/p>\n
                            "\u7279\u65af\u62c9\u63a8\u51fa\u4e86\u5168\u65b0\u7684\u7535\u52a8\u6c7d\u8f66",<\/p>\n
                            "\u8c37\u6b4c\u7684AI<\/a>\u6280\u672f\u53d6\u5f97\u4e86\u91cd\u5927\u7a81\u7834",<\/p>\n
                            "\u5fae\u8f6f\u53d1\u5e03\u4e86\u6700\u65b0\u7684\u64cd\u4f5c\u7cfb\u7edf",<\/p>\n
                            "SpaceX\u6210\u529f\u53d1\u5c04\u4e86\u4e00\u9897\u536b\u661f"<\/p>\n
                        ]<\/p>\n
                        <\/code><\/pre>\n<\/p>\n
                          \n
                        1. \u6587\u672c\u9884\u5904\u7406<\/li>\n<\/ol>\n
                          \u5bf9\u65b0\u95fb\u6587\u7ae0\u8fdb\u884c\u5206\u8bcd\u3001\u53bb\u505c\u7528\u8bcd\u548c\u8bcd\u5e72\u5316\u5904\u7406\u3002<\/p>\n<\/p>\n
                          import jieba<\/p>\n
                          from nltk.corpus import stopwords<\/p>\n
                          from nltk.stem import PorterStemmer<\/p>\n
                          \u5206\u8bcd<\/strong><\/h2>\n
                          tokenized_articles = [jieba.lcut(article) for article in news_articles]<\/p>\n
                          \u53bb\u505c\u7528\u8bcd<\/strong><\/h2>\n
                          stop_words = set(stopwords.words('english'))<\/p>\n
                          filtered_articles = [[word for word in article if word not in stop_words] for article in tokenized_articles]<\/p>\n
                          \u8bcd\u5e72\u5316<\/strong><\/h2>\n
                          stemmer = PorterStemmer()<\/p>\n
                          stemmed_articles = [[stemmer.stem(word) for word in article] for article in filtered_articles]<\/p>\n
                          <\/code><\/pre>\n<\/p>\n
                            \n
                          1. \u6587\u672c\u8868\u793a<\/li>\n<\/ol>\n
                            \u4f7f\u7528TF-IDF\u5bf9\u65b0\u95fb\u6587\u7ae0\u8fdb\u884c\u5411\u91cf\u5316\u8868\u793a\u3002<\/p>\n<\/p>\n
                            from sklearn.feature_extraction.text import TfidfVectorizer<\/p>\n
                            \u5c06\u5206\u8bcd\u540e\u7684\u6587\u7ae0\u5408\u5e76\u4e3a\u5b57\u7b26\u4e32<\/strong><\/h2>\n
                            joined_articles = [" ".join(article) for article in stemmed_articles]<\/p>\n
                            \u8fdb\u884cTF-IDF\u5411\u91cf\u5316<\/strong><\/h2>\n
                            vectorizer = TfidfVectorizer()<\/p>\n
                            tfidf_matrix = vectorizer.fit_transform(joined_articles)<\/p>\n
                            <\/code><\/pre>\n<\/p>\n
                              \n
                            1. \u6587\u672c\u805a\u7c7b<\/li>\n<\/ol>\n
                              \u4f7f\u7528K-means\u5bf9\u65b0\u95fb\u6587\u7ae0\u8fdb\u884c\u805a\u7c7b\u3002<\/p>\n<\/p>\n
                              from sklearn.cluster import KMeans<\/p>\n
                              \u521d\u59cb\u5316KMeans\u805a\u7c7b\u6a21\u578b<\/strong><\/h2>\n
                              kmeans = KMeans(n_clusters=3, random_state=42)<\/p>\n
                              \u8fdb\u884c\u805a\u7c7b<\/strong><\/h2>\n
                              kmeans.fit(tfidf_matrix)<\/p>\n
                              \u83b7\u53d6\u805a\u7c7b\u7ed3\u679c<\/strong><\/h2>\n
                              labels = kmeans.labels_<\/p>\n
                              print(labels)<\/p>\n
                              <\/code><\/pre>\n<\/p>\n
                                \n
                              1. \u805a\u7c7b\u7ed3\u679c\u5206\u6790<\/li>\n<\/ol>\n
                                \u8ba1\u7b97\u805a\u7c7b\u7684\u8f6e\u5ed3\u7cfb\u6570\u548c\u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027\u3002<\/p>\n<\/p>\n
                                from sklearn.metrics import silhouette_score, calinski_harabasz_score<\/p>\n
                                \u8ba1\u7b97\u8f6e\u5ed3\u7cfb\u6570<\/strong><\/h2>\n
                                silhouette_avg = silhouette_score(tfidf_matrix, labels)<\/p>\n
                                print("\u8f6e\u5ed3\u7cfb\u6570:", silhouette_avg)<\/p>\n
                                \u8ba1\u7b97\u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027<\/strong><\/h2>\n
                                ch_score = calinski_harabasz_score(tfidf_matrix.toarray(), labels)<\/p>\n
                                print("\u805a\u7c7b\u5185\u90e8\u4e00\u81f4\u6027:", ch_score)<\/p>\n
                                <\/code><\/pre>\n<\/p>\n
                                \u901a\u8fc7\u4ee5\u4e0a\u6b65\u9aa4\uff0c\u6211\u4eec\u53ef\u4ee5\u4f7f\u7528Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\uff0c\u5e76\u5bf9\u805a\u7c7b\u7ed3\u679c\u8fdb\u884c\u5206\u6790\u3002\u6574\u4e2a\u8fc7\u7a0b\u5305\u62ec\u6587\u672c\u9884\u5904\u7406\u3001\u6587\u672c\u8868\u793a\u3001\u6587\u672c\u805a\u7c7b\u548c\u805a\u7c7b\u7ed3\u679c\u5206\u6790\u3002\u4e0d\u540c\u7684\u805a\u7c7b\u7b97\u6cd5\u9002\u7528\u4e8e\u4e0d\u540c\u7684\u6570\u636e\u96c6\u548c\u5e94\u7528\u573a\u666f\uff0c\u53ef\u4ee5\u6839\u636e\u5177\u4f53\u9700\u6c42\u9009\u62e9\u5408\u9002\u7684\u805a\u7c7b\u65b9\u6cd5\u3002<\/p>\n<\/p>\n
                                \u76f8\u5173\u95ee\u7b54FAQs\uff1a<\/strong><\/h2>\n
                                 \u5982\u4f55\u9009\u62e9\u5408\u9002\u7684\u805a\u7c7b\u7b97\u6cd5\u6765\u5904\u7406\u5206\u8bcd\u540e\u7684\u6587\u672c\u6570\u636e\uff1f<\/strong>
                                \u9009\u62e9\u805a\u7c7b\u7b97\u6cd5\u65f6\uff0c\u53ef\u4ee5\u8003\u8651\u6570\u636e\u7684\u7279\u70b9\u548c\u9700\u6c42\u3002\u5e38\u89c1\u7684\u7b97\u6cd5\u5305\u62ecK-means\u3001\u5c42\u6b21\u805a\u7c7b\u548cDBSCAN\u7b49\u3002K-means\u9002\u5408\u5904\u7406\u5927\u89c4\u6a21\u6570\u636e\uff0c\u6613\u4e8e\u5b9e\u73b0\uff0c\u4f46\u9700\u8981\u9884\u5148\u6307\u5b9a\u805a\u7c7b\u6570\uff1b\u5c42\u6b21\u805a\u7c7b\u5219\u9002\u5408\u5bf9\u6570\u636e\u7684\u5c42\u6b21\u5173\u7cfb\u8fdb\u884c\u5206\u6790\uff1bDBSCAN\u5728\u5904\u7406\u566a\u58f0\u6570\u636e\u65f6\u8868\u73b0\u826f\u597d\u3002\u901a\u8fc7\u5c1d\u8bd5\u4e0d\u540c\u7b97\u6cd5\u5e76\u8bc4\u4f30\u7ed3\u679c\uff0c\u53ef\u4ee5\u627e\u5230\u6700\u9002\u5408\u7684\u805a\u7c7b\u65b9\u6cd5\u3002<\/p>\n
                                \u5728\u8fdb\u884c\u6587\u672c\u805a\u7c7b\u4e4b\u524d\uff0c\u5982\u4f55\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u9884\u5904\u7406\uff1f<\/strong>
                                \u9884\u5904\u7406\u662f\u6587\u672c\u805a\u7c7b\u7684\u91cd\u8981\u6b65\u9aa4\u3002\u53ef\u4ee5\u901a\u8fc7\u53bb\u9664\u505c\u7528\u8bcd\u3001\u8bcd\u5e72\u63d0\u53d6\u6216\u8bcd\u5f62\u8fd8\u539f\u6765\u51cf\u5c11\u566a\u58f0\u3002\u6b64\u5916\uff0c\u4f7f\u7528TF-IDF\u6216\u8bcd\u5411\u91cf\uff08\u5982Word2Vec\u3001GloVe\u7b49\uff09\u5c06\u6587\u672c\u8f6c\u6362\u4e3a\u6570\u503c\u5f62\u5f0f\uff0c\u6709\u52a9\u4e8e\u63d0\u9ad8\u805a\u7c7b\u6548\u679c\u3002\u786e\u4fdd\u6570\u636e\u683c\u5f0f\u4e00\u81f4\u6027\u4e5f\u662f\u9884\u5904\u7406\u7684\u5173\u952e\u73af\u8282\uff0c\u8fd9\u6837\u6709\u52a9\u4e8e\u7b97\u6cd5\u66f4\u597d\u5730\u7406\u89e3\u6570\u636e\u3002<\/p>\n
                                \u5982\u4f55\u8bc4\u4f30\u805a\u7c7b\u7ed3\u679c\u7684\u6548\u679c\uff1f<\/strong>
                                \u8bc4\u4f30\u805a\u7c7b\u6548\u679c\u53ef\u4ee5\u4f7f\u7528\u591a\u79cd\u6307\u6807\u3002\u5e38\u7528\u7684\u5185\u90e8\u8bc4\u4ef7\u6307\u6807\u5305\u62ec\u8f6e\u5ed3\u7cfb\u6570\u548cDavies-Bouldin\u6307\u6570\uff0c\u8fd9\u4e9b\u6307\u6807\u5e2e\u52a9\u8861\u91cf\u805a\u7c7b\u7684\u7d27\u51d1\u6027\u548c\u5206\u79bb\u5ea6\u3002\u6b64\u5916\uff0c\u5916\u90e8\u8bc4\u4ef7\u6307\u6807\u5982\u8c03\u6574\u540e\u7684\u5170\u5fb7\u6307\u6570\uff08Adjusted Rand Index\uff09\u53ef\u4ee5\u4e0e\u771f\u5b9e\u6807\u7b7e\u8fdb\u884c\u6bd4\u8f83\u3002\u901a\u8fc7\u8fd9\u4e9b\u8bc4\u4ef7\u65b9\u6cd5\uff0c\u53ef\u4ee5\u66f4\u5168\u9762\u5730\u7406\u89e3\u805a\u7c7b\u7684\u8d28\u91cf\u548c\u6548\u679c\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"Python\u5bf9\u5206\u8bcd\u540e\u7684\u6587\u672c\u8fdb\u884c\u805a\u7c7b\u7684\u65b9\u6cd5\u6709\u5f88\u591a\u79cd\uff0c\u4e3b\u8981\u5305\u62ecTF-IDF\u5411\u91cf\u5316\u3001K-means\u805a\u7c7b\u3001DBSCAN […]","protected":false},"author":3,"featured_media":1117208,"comment_status":"closed","ping_status":"","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[37],"tags":[],"acf":[],"_links":{"self":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1117200"}],"collection":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/comments?post=1117200"}],"version-history":[{"count":"1","href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1117200\/revisions"}],"predecessor-version":[{"id":1117213,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/posts\/1117200\/revisions\/1117213"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media\/1117208"}],"wp:attachment":[{"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/media?parent=1117200"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/categories?post=1117200"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/docs.pingcode.com\/wp-json\/wp\/v2\/tags?post=1117200"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}