摘要
为解决协同过滤技术中存在的稀疏性、可扩展性问题,提出了一种基于自适应人工免疫网络的协同过滤算法。该算法将协同过滤推荐技术与自适应人工免疫网络相结合,利用人工免疫网络自身的克隆变异机制产生隐式评价来降低数据稀疏性,利用克隆抑制、网络抑制机制减少数据维度来提高可扩展性。实验结果表明,该算法提高了推荐精度,具有一定的实际意义。
To deal with the problems of data sparsity and scalability which exist in the collaborative filtering technology, a new collaborative filtering algorithm based on adaptive artificial immune network is presented. The collaborative filtering technology and adaptive artificial immune network are combined so that the algorithm can take the advantage of the colne and mutation of the artificial immune network to generate implicit ratings to reduce the data sparsity and use the clone suppression and network suppression to reduce the data dimension to improve scalability. The experimental results show that the algorithm improve the recommend accuracy and have some practical significance.
出处
《计算机工程与设计》
CSCD
北大核心
2010年第5期1042-1044,1107,共4页
Computer Engineering and Design
基金
北京市教委基金项目(KM2008100028018)
关键词
协同过滤
自适应人工免疫网络
推荐系统
稀疏性
可扩展性
collaborative filtering
adaptive artificial immune network
recommender system
sparsity
scalability