摘要
基于二分网络优化的推荐算法以物质扩散算法和热传导算法为基础,但是到目前为止,此类算法并未充分考虑到用户在对资源选择偏好的时间效应,而时间效应则是推荐算法中需要考虑的重要因素之一.针对这一问题,提出一种融合物质扩散热传导和时间效应的推荐算法,首先将物质扩散算法和热传导算法混合使用,然后在此基础上,根据用户对资源选择偏好主要受近期选择资源的影响,同时用户的兴趣也会有一定的保留,分别引入两个调节因子增加用户对资源选择偏好的时间效应.实验结果表明,所提出的算法在推荐的准确性和多样性方面都有显著的提高.
The proposed algorithm based on bipartite network optimization is based on the mass diffusion algorithm and the heat conduction algorithm,but so far,the algorithm does not take full account of the user′s time effect on the resource selection preference,and the time effect is the recommendation algorithm one of the important factors.In order to solve this problem,this paper proposes a recommendation algorithm fusing mass diffusion heat conduction and time effect.Firstly,the mass diffusion algorithm and the heat conduction algorithm are mixed together.Then,based on the user′s preference for resource selection,At the same time,the user′s interest will also be preserved,and two adjustment factors are introduced to increase the user′s preference effect on resource selection.Experimental results show that the proposed algorithm has a significant improvement in the accuracy and diversity.
出处
《小型微型计算机系统》
CSCD
北大核心
2017年第9期2056-2061,共6页
Journal of Chinese Computer Systems
基金
国家自然科学基金项目(61373148
61502151)资助
山东省自然基金项目(ZR2014FL010)资助
山东省社科规划项目(2012BXWJ01
15CXWJ13
16CFXJ05)资助
关键词
二分网络
物质扩散
热传导
时间效应
推荐算法
bipartite network
mass diffusion
heat conduction
time effect
recommendation algorithm