文本相似度计算方法

编辑距离

def edit_distance(str1, str2):
    len1 = len(str1)
    len2 = len(str2)
    # 初始化矩阵
    matrix = [[i + j for j in range(len2 + 1)] for i in range(len1 + 1)]
    for row in range(len1):
        for col in range(len2):
            comp = [matrix[row + 1][col] + 1, matrix[row][col + 1] + 1]
            if str1[row] == str2[col]:
                comp.append(matrix[row][col])
            else:
                comp.append(matrix[row][col] + 1)
            # 对相邻字符交换位置的处理判断
            if row > 0 and col > 0:
                if str1[row] == str2[col - 1] and str1[row - 1] == str2[col]:
                    comp.append(matrix[row - 1][col - 1] + 1)
            matrix[row + 1][col + 1] = min(comp)
    return matrix[len1][len2]