用python实现中文相似度匹配算法

以下是Python实现中文相似度匹配算法的示例代码：

1. 余弦相似度算法

import jieba
import numpy as np

def cosine_similarity(s1, s2):
    # 对两个文本分词
    words1 = jieba.cut(s1)
    words2 = jieba.cut(s2)
    # 将分词结果转化为set
    words_set = set(words1).union(set(words2))
    # 构建文本向量
    v1 = np.zeros(len(words_set))
    v2 = np.zeros(len(words_set))
    i = 0
    word_dict = {}
    for word in words_set:
        word_dict[word] = i
        i += 1
    for word in words1:
        v1[word_dict[word]] += 1
    for word in words2:
        v2[word_dict[word]] += 1
    # 计算余弦相似度
    return np.dot(v1, v2) / (np.linalg.norm(v1) * np.linalg.norm(v2))

2. Jaccard相似度算法

import jieba

def jaccard_similarity(s1, s2):
    # 对两个文本分词
    words1 = set(jieba.cut(s1))
    words2 = set(jieba.cut(s2))
    # 计算交集和并集
    intersection = words1 & words2
    union = words1 | words2
    # 计算Jaccard相似度
    return len(intersection) / len(union)

3. 编辑距离算法

def edit_distance(s1, s2):
    m, n = len(s1), len(s2)
    # 定义二维数组dp
    dp = [[0] * (n + 1) for _ in range(m + 1)]
    # 初始化dp
    for i in range(m + 1):
        dp[i][0] = i
    for j in range(n + 1):
        dp[0][j] = j
    # 动态规划计算编辑距离
    for i in range(1, m + 1):
        for j in range(1, n + 1):
            if s1[i - 1] == s2[j - 1]:
                dp[i][j] = dp[i - 1][j - 1]
            else:
                dp[i][j] = min(dp[i - 1][j], dp[i][j - 1], dp[i - 1][j - 1]) + 1
    # 返回编辑距离
    return dp[m][n]

4. 词向量匹配算法

import jieba
import gensim

def word2vec_similarity(s1, s2):
    # 加载预训练的词向量模型
    model = gensim.models.KeyedVectors.load_word2vec_format('pretrained_word2vec.bin', binary=True)
    # 对两个文本分词
    words1 = jieba.cut(s1)
    words2 = jieba.cut(s2)
    # 将分词结果转化为set
    words_set = set(words1).union(set(words2))
    # 计算两个文本的词向量之间的相似度
    v1 = np.zeros(300)
    v2 = np.zeros(300)
    for word in words1:
        if word in model:
            v1 += model[word]
    for word in words2:
        if word in model:
            v2 += model[word]
    return np.dot(v1, v2) / (np.linalg.norm(v1) * np.linalg.norm(v2))

以上是一些常见的中文相似度匹配算法的Python实现示例代码。需要注意的是，对于词向量匹配算法，需要提前下载预训练好的词向量模型（如Word2Vec模型），并使用gensim库来加载和使用这些模型。