基于用户模糊聚类的两阶段协同过滤推荐

推荐系统中,随着用户数目和商品数目的日益增加,传统的协同过滤技术在生成推荐时的速度已经成为一种瓶颈。针对此问题,本文提出了一种基于用户模糊聚类的两阶段协同过滤推荐。两阶段分为离线和在线两个阶段。离线时,应用模糊聚类技术,对基本用户进行模糊聚类;在线时,利用已有的用户模糊聚类寻找目标用户的最近邻居,并产生推荐。实验表明,基于用户模糊聚类的两阶段协同过滤推荐不仅加快了推荐生成速度,还提高了推荐质量。

Two-Phase Collaborative Filtering Recommendation Based on User's Fuzzy Clustering

The magnitudes of users and commodities grow rapidly,and result in the difficulty of the speed bottleneck of collaborative filtering.A two-phase recommendation based on fuzzy clustering is presented to solve this problem.It separates the procedure of recommendation into offline and online phases.In the offline phase,the basic users are clustered using fuzzy clustering;while in the online phase,the nearest neighbors of an active user are found according to the basic users' clusters,and the recommendation to ...