多层聚簇中基于协同过滤的跨类推荐算法

多层聚簇中基于协同过滤的跨类推荐算法

李瑞远;洪亮;曾承

【期刊名称】《小型微型计算机系统》 【年(卷),期】2017(038)004

【摘要】Most electronic commerce (E-commerce ） systems use Collaborative Filtering (CF)-based methods to recommend items belonging to different categories.Market basket analysis has found that a group of like-minded users have similar tastes on items belonging to a subset of correlated categories(called cross-category dependence） rather than all the categories.Therefore,we should consider both user-to-user similarity and item-to-item similarity in recommendations.In real applications,items are usually organized into a multi-level taxonomy which provides hierarchical relationships between items and categories.Note that,the degree of data sparsity varies in different level of categories as there are more items in a category than those in any of its sub-categories.To alleviate the data sparsity problem of existing recommendation

methods,we

propose

an

efficient

multi-level

biclustering algorithm to mine user-iterr/category biclusters (i.e.cross-category dependence ） at each level of the taxonomy.Then we propose a general framework for cross-category recommendation which extends existing CF methods by utilizing multi-level biclusters to improve their recommendation performance.Experiments on a real datasets show that