摘要
【目的】为了解决传统的协同过滤算法在对用户兴趣建模时出现的推荐准确度下降问题,提出改进算法。【方法】在已有算法的基础上提出一种基于受限随机游走的协同过滤算法,分别构建了基于用户和基于项目的信任网络,通过受限随机游走捕捉并反映用户的兴趣变化及项目流行趋势。【结果】与主流的推荐算法相比,本算法能更有效地捕捉项目流行度和用户兴趣的变化趋势。【结论】本算法避免了盲目游走,降低了推荐误差,提高了推荐准确度。
[Purposes]Traditional collaborative filtering disregards the granularity of users＇ preference drifting and item popularity bias in modeling,thus diminished the accuracy of recommendation.[Methods]A new collaborative filtering algorithm is proposed based on Restricted Random Walk.Two new trust network：user-based and item-based are proposed,with Restricted Random Walk to adaptively track the change of users＇ preference drifting and item popularity bias.[Findings]Experimental results on social dataset show that the proposed algorithm could capture the popularity of items and users＇ preference drifting compared with other algorithms.[Conclusions]The proposed algorithm avoids blind walking,reduces the recommendation error and improves the accuracy of recommendation.
作者
陈斌
CHEN Bin(College of Tourism and Culture,Yunnan University,Lijiang Yunnan 674100,China)
出处
《重庆师范大学学报（自然科学版）》
CAS
CSCD
北大核心
2018年第6期82-87,共6页
Journal of Chongqing Normal University:Natural Science
基金
云南省教育厅自然科学基金(No.2015Y358)
关键词
社会网络
协同过滤
随机游走
云相似
social networks
collaborative filtering
random walk
cloud similarity
