基于用户的协作过滤信息推荐模型研究

Study on recommendation model of collaborative filtering based on user

当网络成为人们获取信息的主要途径时,"信息过量"与"信息饥饿"的矛盾却日益凸现,因此,提供个性化服务显得尤为必要。提出了一种基于用户的协作过滤信息推荐模型,实验结果表明,该模型能够有效地改善传统协作过滤推荐技术所面临的扩展性和数据高维稀疏性问题,同时信息推荐质量较传统推荐算法还有明显提高。

Nowadays, web has become the main ways to gain information. However, ＂information overload＂ and ＂information lack＂ has become a big problem to be studied. To provide the personalized service for the people appears especially essential. A recommendation model of collaborative filtering based on user is proposed. The results of experiment show that the model improve the two problems that traditional collaborative filtering faced efficiently. Simultaneously the quality of information recommendation also has the distinct enhancement compares to the traditional recommendation.