基于属性值偏好矩阵的协同过滤推荐算法

传统的协同过滤推荐算法面临用户评分数据稀疏性和冷启动问题的挑战.针对上述问题,提出了基于属性值偏好矩阵的协同过滤推荐算法,首先采用奇异值分解(SVD)对用户-项目评分矩阵降维得到目标用户的初始邻居用户集,生成新的用户-项目评分矩阵;然后将用户评分映射到相应的项目属性值上,生成每个用户的属性值偏好矩阵,并基于属性值偏好矩阵进行用户相似性度量,从而缓解了评分数据稀疏性;将新项目的属性值与用户的属性值偏好矩阵进行匹配,从而找出匹配度最高的前N个用户作为新项目的推荐受众.实验结果表明了该算法的有效性.

A Collaborative Filtering Recommendation Algorithm Based on Attributes-value Preference Matrix

Traditional collaborative filtering recommendation algorithm faces the challenge of sparse user ratings and cold- start problem.A collaborative filtering recommendation algorithm based on Attributes-value Preference Matrix(APM)is proposed to solve the problem.At first Singular Value Decomposition(SVD)has been used to reduce the dimensionality of user-item rating matrix,thus the initial neighbor set for target user can be gained,and a new user-item rating matrix will be created.Then user ratings have been ma...