论文部分内容阅读
根据目前电子商务网站中商品个性化推荐的现状,重点分析了比较常用的一种推荐方法——协同过滤推荐方法,发现了目前协同过滤算法在应用中所面临的挑战和问题,主要包括:推荐质量低、推荐效率低、数据稀疏性、冷启动等问题。针对这些问题本文提出了一种基于用户兴趣度的聚类分析协同过滤算法,有效的解决了目前算法中存在的数据稀疏性等问题,通过实验数据的分析对比,证明了算法的合理性和有效性。
According to the status quo of product personalization recommendation in e-commerce websites at present, this paper mainly analyzes one of the more commonly used recommendation methods-collaborative filtering recommendation method and finds out the challenges and problems in the application of collaborative filtering algorithms, including: recommendation Low quality, low recommended efficiency, data sparsity, cold start and other issues. Aiming at these problems, this paper proposes a clustering analysis collaborative filtering algorithm based on user interest, which effectively solves the problem of data sparsity existing in current algorithms. The experimental results show that the algorithm is reasonable and effective Sex.