博客

当然可以，以下是用Python编写的Simhash算法计算文本相似度的代码：

import re
import hashlib

class Simhash(object):
    def __init__(self, content):
        self.simhash = self.get_simhash(content)

    def get_features(self, content):
        # 对文本进行分词，提取出重要的特征词
        features = []
        words = re.compile(u'[\u4e00-\u9fa5]+').findall(content)
        for word in words:
            features.append(hash(word))
        return features

    def get_simhash(self, content):
        features = self.get_features(content)
        # 初始化哈希值列表
        v = [0] * 128
        # 遍历特征词哈希值，进行加权累加
        for feature in features:
            for i in range(128):
                mask = 1 << i
                if feature &amp; mask:
                    v[i] += 1
                else:
                    v[i] -= 1
        # 将加权累加的结果映射到01串上
        simhash = 0
        for i in range(128):
            if v[i] > 0:
                simhash |= 1 << i
        return simhash

    def distance(self, other):
        # 计算两个simhash之间的海明距离
        x = (self.simhash ^ other.simhash) &amp; ((1 << 128) - 1)
        ans = 0
        while x:
            ans += 1
            x &amp;= x - 1
        return ans

    def similarity(self, other):
        # 计算两个simhash之间的相似度
        d = self.distance(other)
        return 1 - d / 128