计算机与现代化

• 数据库与数据挖掘 • 上一篇    下一篇

基于差值矩阵分解的推荐算法

  

  1. (温州大学数理与电子信息工程学院,浙江温州325035)
  • 收稿日期:2017-07-31 出版日期:2018-04-03 发布日期:2018-04-03
  • 作者简介:成鹏(1993-),男,江苏盐城人,温州大学数理与电子信息工程学院硕士研究生,研究方向:数据挖掘,推荐系统; 通信作者:刘文斌(1969-),男,陕西韩城人,教授,博士,研究方向:生物信息学,数据挖掘,DNA计算。
  • 基金资助:
    国家自然科学基金资助项目(60970065,61272018,61572367)

Recommendation Algorithm Based on Difference Value Matrix Factorization

  1. (College of Mathematics, Physics and Electronic Information Engineering, Wenzhou University, Wenzhou 325035, China)
  • Received:2017-07-31 Online:2018-04-03 Published:2018-04-03

摘要: 矩阵分解已经成为预测用户对物品评分的一种常用方法。传统的矩阵分解技术没有考虑到用户评分之间的差异性,针对上述问题在矩阵分解的基础上,提出差值矩阵分解模型。算法将每个用户对物品的评分减去与其社会属性相似用户对该物品评分的平均分,得到一个差值矩阵,然后对差值矩阵进行分解。在Movielens 1M数据集的实验结果表明,该算法的预测精度较贝叶斯概率矩阵分解、矩阵分解、融合用户属性的隐语义模型都有较为明显的提升。

关键词: 推荐算法, 矩阵分解, 差值矩阵分解

Abstract: Matrix factorization has become a common way to predict user ratings of items. Traditional matrix factorization algorithms do not take account of the differences between users. To address this problem, a difference value(D-value) matrix factorization model is proposed. First, for each user, the difference between his/her rating score and the average rating score from users with similar social attributes is calculated, which finally results in a matrix called D-value matrix. Then the D-value matrix is factorized to calculate the predicted ratings. Experimental results on the Movielens 1M dataset show that the proposed method significantly outperforms Bayesian probabilistic matrix factorization, matrix factorization and the latent factor model fused with user attributes in terms of prediction accuracy.

Key words: recommendation algorithms, matrix factorization, D-value matrix factorization

中图分类号: