电子学报 ›› 2018, Vol. 46 ›› Issue (7): 1762-1767.DOI: 10.3969/j.issn.0372-2112.2018.07.032

• 科研通信 • 上一篇    下一篇

一种基于云模型的社交网络推荐系统评分预测方法

肖云鹏, 孙华超, 戴天骥, 李茜, 李暾   

  1. 重庆邮电大学网络与信息安全技术重庆市工程实验室, 重庆 400065
  • 收稿日期:2017-08-07 修回日期:2018-02-10 出版日期:2018-07-25 发布日期:2018-07-25
  • 作者简介:肖云鹏,男.1979年8月出生,安徽蚌埠人.副教授、硕士生导师,主要研究方向为社交网络、机器学习.E-mail:xiaoyp@cqupt.edu.cn;孙华超,男.1989年3月出生,江苏徐州人.现为重庆邮电大学硕士研究生,主要研究方向为推荐系统.
  • 基金资助:
    国家973重点基础研究发展计划(No.2013CB329606);国家自然科学基金(No.61772098);重庆市重点研发项目(No.cstc2017zdcy-zdyf0299,No.cstc2017zdcy-zdyf0436);重庆市基础科学与前沿技术研究项目(No.cstc2017jcyjAX0099)

A Rating Prediction Method Based on Cloud Model in Social Recommendation System

XIAO Yun-peng, SUN Hua-chao, DAI Tian-ji, LI Qian, Li Tun   

  1. Chongqing Engineering Laboratory of Internet and Information Security, Chongqing University of Posts and Telecommunications, Chongqing 400065, China
  • Received:2017-08-07 Revised:2018-02-10 Online:2018-07-25 Published:2018-07-25

摘要: 本文针对评分预测中用户评分主观性及评分数据稀疏带来的预测不准确问题,围绕社交推荐的特点,设计实现一种社交网络评分预测方法.首先,针对评分主观性问题,引入并优化相关云模型理论,提出采用综合云模型生成评分标准并转化用户评分的方法.其次,针对预测不准确问题,通过引入隶属度达到数据降维和目标用户定位的作用,同时考虑到社交网络用户关系对评分结果的影响,分别利用社交关系及相似群体建立两个评分预测模型,并基于高斯变换融合两部分预测结果生成预测评分.实验表明,该方案不仅克服了用户评分主观性,同时有效改善了用户评分数据稀疏情况下传统预测方法准确度偏差的问题.

关键词: 社交网络, 推荐系统, 评分预测, 云模型

Abstract: Focusing on the issues of rating prediction such as subjectivity of user rating and inaccurate prediction caused by rating sparsity,a rating prediction method is proposed by introducing the characteristics of social recommendation.Firstly,aiming at the subjective of user rating,we introduce and optimize the cloud model theory.Then,a method to generate rating standard by synthetical cloud model and transform user rating under the standard is proposed.Secondly,to deal the problem of inaccurate score prediction caused by data sparseness,data dimension reduction and target user location are achieved by introducing membership degree.And taking into account that user rating can be affected by their social relationship,we try to learn two rating prediction models by respectively using social relationships and similar groups.Finally,the rating value is obtained by using Gauss transform to combine the two prediction models.Experimental results show that our method not only overcomes subjectivity of user rating,but also alleviates the poor accuracy caused by rating sparsity problem in traditional rating prediction methods.

Key words: social network, recommendation system, rating prediction, cloud model

中图分类号: