基于加权序列模式的推荐算法研究  

Research on the recommender algorithm based on weighted sequential patterns

作　　者：宋威[1] 乔阳阳 

机构地区：[1]北方工业大学计算机学院,北京100144

出　　处：《计算机工程与科学》2015年第7期1399-1404,共6页Computer Engineering & Science

基　　金：国家自然科学基金资助项目(61105045;51075423);北方工业大学科研人才提升计划项目(CCXZ201303)

摘　　要：由于考虑了用户的访问顺序,基于序列模式的推荐方法正在成为推荐系统研究的热点之一。为提高推荐结果的个性化程度,提出了一种基于加权序列模式的推荐算法PRWSP。首先,给出了新的加权序列模式模型,该模型在设置权重时充分考虑了项目在不同序列中的不同重要程度。其次,通过近似估计序列权重的方式,论证了挖掘加权序列模式时同样满足反单调性,从而约简了搜索空间。最后,定义了序列模式匹配程度的度量标准。实验结果表明,PRWSP算法具有较高的挖掘效率和推荐精度。Considering users＇ access order, the recommendation approach based on sequential pat- terns is becoming one hot topic in the field of recommender system. To improve the level of personaliza- tion, we propose a recommender algorithm named personalized recommendation based on weighted se- quential patterns（PRWSP）. We first present a new weighted sequential pattern model, in which the dif- ferent importance degrees of the items in different sequences are considered. Furthermore, by approxi- mation, the rationale of anti-monotonicity in mining weighted sequential patterns is discussed, thus the searching space is reduced. Finally, the measurement metrics of the matching degree of the sequential patterns are defined. Experimental results show that the PRWSP algorithm has higher mining efficiency and recommendation accuracy.

关 键 词：数据挖掘 加权序列模式 反单调性 推荐算法 

分 类 号：TP311[自动化与计算机技术—计算机软件与理论]