Scikit学习K-means聚类和TfidfVectorizer:如何将tf-idf得分最高的前n个术语传递给k-means

from sklearn.feature_extraction.text import TfidfVectorizer
from sklearn.cluster import KMeans

def kmeans_clustering(data, n_clusters, n_top_terms):
    # 创建TfidfVectorizer对象
    vectorizer = TfidfVectorizer()
    
    # 计算文本数据的TF-IDF矩阵
    tfidf_matrix = vectorizer.fit_transform(data)
    
    # 获取TF-IDF得分最高的前n个术语
    top_terms = get_top_terms(tfidf_matrix, vectorizer.get_feature_names(), n_top_terms)
    
    # 执行K-means聚类
    kmeans = KMeans(n_clusters=n_clusters)
    kmeans.fit(tfidf_matrix)
    
    # 获取聚类结果
    clusters = kmeans.labels_
    
    return top_terms, clusters

def get_top_terms(tfidf_matrix, feature_names, n_top_terms):
    # 获取TF-IDF得分最高的前n个术语
    top_terms = []
    for i in range(tfidf_matrix.shape[0]):
        row = tfidf_matrix.getrow(i).toarray()[0]
        top_indices = row.argsort()[-n_top_terms:][::-1]
        top_terms.append([feature_names[j] for j in top_indices])
    
    return top_terms

data = ["文本数据1", "文本数据2", "文本数据3", ...]  # 替换为实际的文本数据
n_clusters = 3  # 聚类数
n_top_terms = 5  # 前n个术语

top_terms, clusters = kmeans_clustering(data, n_clusters, n_top_terms)

# 打印得分最高的术语和聚类结果
for i in range(len(data)):
    print("文本数据{}的得分最高的术语：{}".format(i+1, top_terms[i]))
    print("文本数据{}的聚类结果：{}".format(i+1, clusters[i]))