网络电视推荐系统框架及协同过滤算法的研究

针对网络电视推荐系统中通常采用的协同过滤推荐算法的不足,提出了一种将聚类、用户相似—信任关系和项目属性关系相组合的协同过滤推荐技术.该组合推荐技术首先通过聚类分析缩小用户的有效搜索范围,其次通过引入信任关系来提高推荐的准确性,从而为目标用户提供更好的推荐结果.经过实验表明,该算法提高了推荐质量.

Network TV Recommended System Framework and Research of Collaborative Filtering Algorithm

In view of the Insufficient of network TV recommender system which usually uses collaborative filtering recommendation algorithms, put forward a kind of combination of recommendation algorithm which is based on Clustering Similarty-Trust and Project properties relationship. Firstly,the combination of recommendation algorithm reduced user's effective range through cluster analysis,then improve the accuracy of recommendation through the trust mechanism, so as to provide better recommendation results for target users. The experimental results show that, the algorithm improves the quality of recommendation.