| 摘 要: 社交网络发展迅猛,社会网络环境下的信息量骤增,如何在大数据下向用户推荐感兴趣的项目是当前研 究的热点问题之一。目前的推荐系统在用户反馈数据稀疏的情况下和向新用户推荐中存在推荐不准确的问题,为了提高 推荐质量,提出了一种融合社会标签的联合概率矩阵分解推荐模型TaSoRec,该模型运用社交网络的用户、项目、标签 三者信息进行推荐,通过对训练模型参数优化,从而提升推荐效果。 |
| 关键词: 社会标签;联合概率矩阵;推荐方法;社交网络 |
|
中图分类号: TP181
文献标识码: A
|
|
| Incorporating Social Tagging for Uni ed Probabilistic Matrix Factorization Recommendation |
|
ZHENG Shaozhen1, ZHENG Dongxia2
|
( 1. Liaoning Jinyang Group Information Technology Co., Ltd ., Dandong 118000, China ;
2.School of Computer and Software, Dalian Neusoft University of Information, Dalian 116023, China)
94225621@qq.com; zhengdongxia@neusoft.edu.cn
|
| Abstract: With the development of social network, the explosive growth of information makes a hot issue to recommend attractive resources to users based on big data. In order to improve the quality of recommendation, this paper proposes a joint probability matrix factorization recommendation model called TaSoRec (Tag Social Recommendation) which integrates social tags. The model uses information of users, resources and tags of social network, and optimizes the parameters of the training model to achieve better recommendation results. |
| Keywords: social tagging; unified probability matrix; recommendation method; social network |
