基于上下文学习和张量分解的个性化推荐  

PERSONALIZED RECOMMENDATION BASED ON CONTEXT LEARNING AND TENSOR DECOMPOSITION

作　　者：刘振娇 王新华[1,2] 

机构地区：[1]山东师范大学信息科学与工程学院,济南250014 [2]山东省分布式计算机软件新技术重点实验室,济南250014

出　　处：《山东师范大学学报（自然科学版）》2017年第1期36-42,共7页Journal of Shandong Normal University(Natural Science)

基　　金：国家自然科学基金资助项目(9061200;61572301);山东省自然科学基金资助项目(ZE2013FM008)

摘　　要：针对目前推荐系统存在的用户评分稀疏性现象,该文提出了一种基于上下文学习和张量分解的个性化推荐算法,算法通过利用用户之间共同评价的项目的上下文信息与评价过项目的用户上下文信息分别构建两个三阶张量,并应用高阶奇异值分解充分挖掘上述两个三阶张量实体之间潜在的关联关系,并将张量分解后的两个三维张量进行组合进而得到最终的推荐列表,以响应用户个性化请求.实验结果表明,该算法可以有效地对上下文信息进行建模,可以显著提高在数据稀疏情况下的推荐质量.To solve the user rating sparsity problem existing in present recommender systems,this paper proposed a context learning personalized recommendation algorithm with tensor decomposition.By using the context information of the project commented by users and the users who have commented the project,algorithm manages to create two third-order susceptibility respectively.And then,we use a high order singular value decomposition method to mine the potential semantic association of the above two third-order.Finally,the resulted tensors were combined to reach the recommendation list to respond the users＆#39; personalized query requests.experimental results show that the proposed algorithm can effectively improve the effectiveness of the recommendation system.Especially in the case of sparse data,and it can significantly improve the quality of the recommendation.

关 键 词：个性化推荐 上下文感知 张量分解 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]