k-means文本聚类算法代码

k-means是一种常用的文本聚类算法，其基本思想是将文本数据集分成k个不同的簇，每个簇都具有相似的特征。具体实现时，首先需要随机选择k个数据点作为初始质心，然后将每个数据点分配到距离其最近的质心所在的簇中，接着重新计算每个簇的质心位置，重复上述过程直到收敛。

下面是k-means文本聚类算法的Python代码实现：

import numpy as np
from sklearn.cluster import KMeans
from sklearn.feature_extraction.text import TfidfVectorizer

def kmeans_cluster(docs, num_clusters):
    # 提取文本特征
    vectorizer = TfidfVectorizer(stop_words='english')
    X = vectorizer.fit_transform(docs)

    # 聚类
    kmeans = KMeans(n_clusters=num_clusters, random_state=0)
    kmeans.fit(X)

    # 输出聚类结果
    clusters = [[] for i in range(num_clusters)]
    for i, label in enumerate(kmeans.labels_):
        clusters[label].append(i)
    for i, cluster in enumerate(clusters):
        print('Cluster %d:' % i)
        for doc_index in cluster:
            print('\t%s' % docs[doc_index])

# 示例
docs = ['This is a sample sentence.',
        'Another example sentence.',
        'I love programming.',
        'Python is a great programming language.',
        'Java is also a popular programming language.']
kmeans_cluster(docs, 2)

运行结果如下：

Cluster 0:
        I love programming.
        Python is a great programming language.
        Java is also a popular programming language.
Cluster 1:
        This is a sample sentence.
        Another example sentence.

以上代码通过TfidfVectorizer提取文本特征，并使用KMeans聚类算法进行文本聚类。