item-based CF(基于物品的协同过滤算法)

最新推荐文章于 2025-05-19 16:54:25 发布

pb_0219

最新推荐文章于 2025-05-19 16:54:25 发布

阅读量1.5k

点赞数

CC 4.0 BY-SA版权

分类专栏： recommendation 文章标签：协同过滤 python CF 推荐 recommendation

本文链接：https://blog.youkuaiyun.com/qq_21944445/article/details/84887336

本文详细介绍了基于物品的协同过滤（Item-Based CF）算法，包括其原理、实现过程和应用。通过学习，读者将能够理解如何使用Python进行推荐系统的开发。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

#!/usr/bin/env python
# -*- coding: utf-8 -*-

"""
item_based CF
"""


import numpy as np
import pandas as pd


class CFItem(object):
    def __init__(self, k):
        self.top_k = k

    @staticmethod
    def simu_cal_item(data_beha, i, j):
        data_beha = np.array(data_beha)
        data_beha = data_beha[:, [i, j]]
        index = np.where(data_beha == 0)[0]
        data_beha_a = np.delete(data_beha, obj=index, axis=0)
        if data_beha_a.size == 0:
            return 0
        else:
            data_beha_i = data_beha_a[:, 0]
            data_beha_j = data_beha_a[:, 1]
            return np.corrcoef(x=data_beha_i, y=data_beha_j)[0, 1]

    def get_simu_mat(self, data_beha):
        data_beha = np.array(data_beha)
        mat = np.zeros([data_beha.shape[1], data_beha.shape[1]])
        for i in range(mat.shape[0]):
            for j in range(mat.shape[1]):
                mat[i][j] = self.simu_cal_item(data_beha, i, j)
                mat[j][i] = mat[i][j]