广东工业大学学报 ›› 2019, Vol. 36 ›› Issue (03): 39-46.doi: 10.12052/gdutxb.180112

• 综合研究 • 上一篇    下一篇

结合强弱联系和兴趣的社交网络推荐算法

何炜俊1, 周应堂2   

  1. 1. 广东工业大学 管理学院, 广东 广州 510520;
    2. 南京农业大学 工学院, 江苏 南京 210031
  • 收稿日期:2018-08-16 出版日期:2019-05-09 发布日期:2019-04-04
  • 通信作者: 周应堂(1963-),男,副教授,博士,硕士生导师,主要研究方向为管理创新.E-mail:791277872@qq.com E-mail:791277872@qq.com
  • 作者简介:何炜俊(1995-),男,硕士研究生,主要研究方向为推荐系统.
  • 基金资助:
    国家自然科学基金资助项目（71740024，71472036）；江苏高校哲学社会科学重点项目（2017ZDIXM015）

A Social Network Recommendation Algorithm Combining Strong and Weak Ties and Interests

He Wei-jun1, Zhou Ying-tang2   

  1. 1. School of Management, Guangdong University of Technology, Guangzhou 510520, China;
    2. College of Engineering, Nanjing Agricultural University, Nanjing 210031, China
  • Received:2018-08-16 Online:2019-05-09 Published:2019-04-04

HTML

PDF (PC)

2721

摘要： 提出了一种结合强弱联系和兴趣的社交网络推荐算法.首先，考虑依据强弱联系为客户构建社交关系集合，同时兼顾信息传递的广度和深度.然后，基于关联规则改进传统PageRank算法的状态转移概率，修正的矩阵能够更合理地度量不同客户之间的社交紧密程度.同时，考虑客户之间的兴趣爱好相似性，赋予其对候选项目投票的权重，旨在提高系统的多样性和新颖性.最后，综合上述两者对候选项目进行评分并作Top-N过滤得到推荐列表.实验结果表明，本算法相对于参照算法更具合理性和有效性.

关键词: 推荐系统, PageRank算法, 弱联系, Jaccard系数

Abstract: Social network recommender systems combining strong and weak ties and interests are proposed. First, social relationship sets for customers are built based on the strong and weak ties, taking into account the breadth and depth of information transfer. Then, based on the association rules to improve the state transition probability of the PageRank Algorithm, the modified matrix is capable of measuring the social closeness between different customers more reasonably. Meanwhile, aiming to increase diversity and novelty of the systems, the similarity of interests between different customers is considered giving them the weight of voting in the candidate projects. Finally, the candidate items are scored and Top-N filtered to obtain a recommendation list. As is shown in the experiment results, the algorithm is more reasonable and effective than the reference algorithms.

Key words: recommender systems, PageRank algorithm, weak ties, Jaccard index

中图分类号: 

  • C931.6
[1] DAKHEL A M, MALAZI H T, MAHDAVI M. A social recommender system using item asymmetric correlation[J]. Applied Intelligence, 2018, 48(3):527-540
[2] VAN S M. Supporting people in finding information:hybrid recommender systems and goalbased structuring[M]. Netherlands:Enschede, 2005:16.
[3] RECALDE L. A Social Framework for Set Recommendation in Group Recommender Systems[C]//European Conference on In-formation Retrieval. Aberdeen:Springer, 2017:735-743.
[4] YANG X, GUO Y, LIU Y, et al. A survey of collaborative filtering based social recommender systems[J]. Computer Communications, 2014, 41(5):1-10
[5] RICCI F, ROKACH L, SHAPIRA B, et al. Recommender systems handbook[M]. New York:Springer, 2011:511-543.
[6] MA H. An experimental study on implicit social recommendation[C]//Proceedings of the 36th international ACM SIGIR conference on Research and development in information retrieval. Dublin:ACM, 2013:73-82.
[7] GRANOVETTER M S. The strength of weak ties[J]. American Journal of Sociology, 1973, 78(6):1360-1380
[8] RAPOPORT A. Contribution to the theory of random and biased nets[J]. Bulletin of Mathematical Biophysics, 1957, 19(4):257-277
[9] NARUCHITPARAMES J, GÜNEŞ M H, LOUIS S J. Friend recommendations in social networks using genetic algorithms and network topology[C]//Evolutionary Computation (CEC), 2011 IEEE Congress on. New Orleans:IEEE, 2011:2207-2214.
[10] 尹熙成, 朱恒民, 魏静, 等. 在线社交网络中具有传播价值的弱关系识别研究[J]. 情报杂志, 2016, 35(3):125-130 YIN X C, ZHU H M, WEI J, et al. Identifying weak ties with value of dissemination in online social networks[J]. Journal of Intelligence, 2016, 35(3):125-130
[11] ABRAMO G, D'ANGELO C A, SOLAZZI M. The relation-ship between scientists' research performance and the degree of inter-nationalization of their research[J]. Scientometrics, 2011, 86(3):629-643
[12] NAEGELE L, PLATH C. Weak ties get you in the job, strong ties keep you[C]//Zeitschrift Fur Gerontologie Und Geriatrie. Heidelberg:springer, 2018, 51:151-151.
[13] LYU L, ZHOU T. Link prediction in weighted networks:The role of weak ties[J]. Europhysics Letters, 2010, 89(1):18001
[14] ZHU B, XIA Y. Link prediction in weighted networks:A weighted mutual information model[J]. PloS one, 2016, 11(2):e0148265
[15] 张怡文, 岳丽华, 张义飞, 等. 基于共同用户和相似标签的好友推荐方法[J]. 计算机应用, 2013, 33(8):2273-2275 ZHANG Y W, YUE L H, ZHANG Y F, et al. Friends recommended method based on common users and similar labels[J]. Journal of Computer Applications, 2013, 33(8):2273-2275
[16] CHEN Z, JIANG Y, ZHAO Y. A collaborative filtering recommendation algorithm based on user interest change and trust evaluation[J]. Microcomputer & Its Applications, 2010, 4(9):106-113
[17] ZHANG Z, ZHENG X, ZENG D. A framework for diversifying recommendation lists by user interest expansion[J]. Know-ledge-based Systems, 2016, 105(5):83-95
[18] 徐建民, 刘明艳, 王苗. 基于用户扩展兴趣的微博推荐方法[J]. 计算机应用研究, 2019(6):1-6 XU J M, LIU M Y, WANG M. Microblog recommendation method based on extended interest of users[J]. Application Research of Computers, 2019(6):1-6
[19] 李永立, 罗鹏, 张书瑞. 基于决策分析的社交网络链路预测方法[J]. 管理科学学报, 2017, 20(1):64-74 Li Y L, LUO P, ZHANG S R. Link prediction in social networks based on decision analysis[J]. Journal of Management Sciences in China, 2017, 20(1):64-74
[20] SINHA R, SWEARINGEN K. Comparing recommendations made by online systems and friends[C].//In Proceedings of the DELOS-NSF Workshop on Personalization and Recommender Systems in Digital Libraries. Berlin:Springer, 2001.
[21] 郜燕, 刘文芬. 基于隐Markov过程的网络信任评估模型[J]. 四川大学学报(工程科学版), 2015, 47(3):101-107 GAO Y, LIU W F. A dynamic trust evaluation model based on optimized hidden markov process[J]. Journal of Sichuan University (Engineering Science Edition), 2015, 47(3):101-107
[22] 甘春梅, 钟绮桐, 罗婷予. 社会化商务环境下消费者信任形成的影响因素研究[J]. 情报科学, 2017(4):68-73 GAN C M, ZHONG Q T, LUO T Y. Factors affecting trust formation in the context of social commerce[J]. Information Science, 2017(4):68-73
[23] Delicious website, http://www.delicious.com.
[24] NEPAL S, PARIS C, POUR P A, et al. Interaction based con-tent recommendation in online communities[C]//International Conference on User Modeling, Adaptation, and Personalization. Berlin:Springer, 2013:14-24.
[25] ARNABOLDI V, CAMPANA M G, DELMASTRO F, et al. A personalized recommender system for pervasive social networks[J]. Pervasive and Mobile Computing, 2017, 36:3-24
[1] 易闽琦, 刘洪伟, 高鸿铭. 电商平台产品共同购买网络的影响因素研究[J]. 广东工业大学学报, 2022, 39(03): 16-24.
[2] 张巍, 黄健华, 刘冬宁, 滕少华, 刘子婷. 一种改进的结合评分和评论信息的推荐方法[J]. 广东工业大学学报, 2017, 34(06): 27-31,48.
[3] 汪岭, 傅秀芬, 王晓牡. 基于大数据集的混合动态协同过滤算法研究[J]. 广东工业大学学报, 2014, 31(3): 44-48.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
版权所有 © 2017 《广东工业大学学报》编辑部
地址：广州市东风东路729号广东工业大学 邮编：510090
电话：020-37626653,020-37626139,020-37626396
邮箱：xbzrb@gdut.edu.cn
本系统由北京玛格泰克科技发展有限公司设计开发