余弦相似度聚类加速 代码实现

以下是余弦相似度聚类加速的Python代码实现：

import numpy as np

def cosine_similarity(x, y):
    """计算余弦相似度"""
    dot_product = np.dot(x, y)
    norm_x = np.linalg.norm(x)
    norm_y = np.linalg.norm(y)
    return dot_product / (norm_x * norm_y)

def cluster_cosine_similarities(X):
    """计算数据集中所有样本之间的余弦相似度"""
    n_samples = X.shape[0]
    similarities = np.zeros((n_samples, n_samples))
    for i in range(n_samples):
        for j in range(i, n_samples):
            similarity = cosine_similarity(X[i], X[j])
            similarities[i][j] = similarity
            similarities[j][i] = similarity
    return similarities

def cluster(data, threshold):
    """基于余弦相似度聚类"""
    similarities = cluster_cosine_similarities(data)
    n_samples = data.shape[0]
    clusters = []
    visited = set()
    for i in range(n_samples):
        if i not in visited:
            cluster = set()
            for j in range(i, n_samples):
                if similarities[i][j] >= threshold:
                    cluster.add(j)
                    visited.add(j)
            clusters.append(cluster)
    return clusters

使用示例：

data = np.array([
    [1, 2, 3],
    [2, 3, 4],
    [3, 4, 5],
    [4, 5, 6],
    [5, 6, 7],
    [6, 7, 8]
])

clusters = cluster(data, 0.8)
print(clusters)

输出结果：

[{0, 1, 2, 3, 4, 5}]

说明所有样本都被聚类在了一起。