Abstract: Nearest neighbor computation is a typical collaborative filtering approach for high recommendation accuracy. However, this approach is not scalable given the huge number of customers and items in typical commercial applications, cluster-based collaborative filtering techniques can be a remedy for the problem, but they usually provide relatively lower accuracy. This paper provides an efficient implementation of cluster search strategy by adapting a specifically tailored cluster-skipping inverted index structure. Experimental results reveal that this is a good solution for high accuracy and reasonable scalability of the recommender system.

Key words: collaborative filtering recommendation system, clustering search method, inverted index

摘要： 最近邻计算是协同过滤方法中直接影响到推荐系统的运行效率和推荐准确率的重要一环。当用户和项目数目达到一定规模的时候，推荐系统的可扩展性明显降低。聚类的方法能在一定程度上弥补这个缺陷，但同时又会带来推荐准确性的下降。提出了一种与信息检索领域中的倒排索引相结合并采用“成员策略”的用户聚类搜索算法，缩短了最近邻计算的时间，实验的结果证明，该方法能在保证推荐正确性的前提下有效改善协同过滤推荐系统的可扩展性。

关键词: 协同过滤推荐系统, 类搜索方法, 倒排索引