| 用户兴趣对个性化推荐的影响 |
| |
| 引用本文： | 潘新,邓贵仕,刘建国. 用户兴趣对个性化推荐的影响[J]. 控制工程, 2010, 0(Z1) |
| |
| 作者姓名： | 潘新 邓贵仕 刘建国 |
| |
| 作者单位： | 大连理工大学系统工程研究所;上海理工大学复杂系统科学研究中心;弗里堡大学物理系; |
| |
| 基金项目： | 国家自然科学基金资助项目(70972059,70671016,10905052,70901010) |
| |
| 摘 要： | 利用用户-产品二部分图中度量同类节点相似性的加权映射方法,研究了用户的兴趣对基于物质扩散的个性化推荐算法的影响并提出了相应的改进算法,其中用户的兴趣定义为用户所选择过产品的平均度。该算法中在推荐过程中假设用户收集产品被赋予的推荐强度应由用户的兴趣点和产品自身的度一起决定。数值实验结果显示该算法可以提高原物质扩散算法的准确度。进一步,当数据集稀疏的时候,算法应该赋予与用户兴趣点相近的产品更大的推荐强度;随着数据集变得越来越稠密,应该赋予用户的兴趣点外的其他产品更多的权重以提高算法的准确度和推荐列表差异性。
|
| 关 键 词： | 个性化推荐 二部分图 基于网络结构的推荐算法 |
Effects of User Tastes on Personalized Recommendation |
| |
| PAN Xin,DENG Gui-shi,LIU Jian-guo. Effects of User Tastes on Personalized Recommendation[J]. Control Engineering of China, 2010, 0(Z1) |
| |
| Authors: | PAN Xin DENG Gui-shi LIU Jian-guo |
| |
| Affiliation: | 1.Institute of Systems Engineering/a>;Dalian University of Technology/a>;Dalian 116024/a>;China/a>;2.Research Center of Complex Systems Science/a>;University of Shanghai for Science and Technology/a>;Shanghai/a>;2000/a>;3.Department of Physics/a>;University of Fribourg/a>;Fribourg CH-1700/a>;Switzerland |
| |
| Abstract: | Based on a weighted projection of the user-object bipartite network,the effects of user tastes on the mass-diffusion-based personalized recommendation algorithm are studied,where a user tastes or interests are defined by the average degree of the objects he has collected.It is assumed that the initial recommendation power located on the objects should be determined by both of their degree and the users tastes.By introducing a tunable parameter,the user taste effects on the configuration of initial recommend... |
| |
| Keywords: | personalized recornmendation bipartite network network-based recommendation algorithm |
| 本文献已被 CNKI 等数据库收录！ |
