融合链接拓扑结构和用户兴趣的朋友推荐方法

doi:10.11959/j.issn.1000-436x.2015040

通信学报 ›› 2015, Vol. 36 ›› Issue (2): 117-125.doi: 10.11959/j.issn.1000-436x.2015040

融合链接拓扑结构和用户兴趣的朋友推荐方法

尚燕敏^1,²,张鹏^1,²,曹亚男^1,²

¹ 中国科学院计算技术研究所，北京 100190
² 中国科学院信息工程研究所，北京 100093

出版日期:2015-02-25 发布日期:2017-06-27
基金资助:
国家高技术研究发展计划（“863计划）基金资助项目;国家自然科学基金资助项目;国家自然科学基金资助项目

New interest-sensitive and network-sensitive method for user recommendation

Yan-min SHANG^1,²,Peng ZHANG^1,²,Ya-nan CAO^1,²

¹ Institute of Computing Technology,Chinese Academy of Sciences,Beijing 100190,China
² Institute of Information Engineering,Chinese Academy of Sciences,Beijing 100093,China

Online:2015-02-25 Published:2017-06-27
Supported by:
The National High Technology Research and Development Program of China (863 Program);Priority Research Program (XDA06030200); The National Natural Science Foundation of China;Priority Research Program (XDA06030200); The National Natural Science Foundation of China

摘要/Abstract

摘要：

提出一种新的朋友推荐方法，该方法同时使用用户兴趣和朋友关系这2种因素来为目标用户推荐朋友，对PageRank算法进行改进，提出一种能同时融合上述2种因素的Topic_Friend_PageRank(TFPR)模型。首先，采用LDA(latent Dirichlet allocation)分析用户发布的消息内容，将用户表示为若干主题上的分布，从而建模用户的兴趣。接下来，使用加权的 PageRank 算法建模用户在整个链接拓扑中的重要程度和用户之间朋友关系的相似性。最后根据主题感知的PageRank思想，将用户兴趣融入前面提到的加权PageRank中，形成同时融合用户兴趣和朋友关系的TFPR模型。采用新浪微博数据验证所提模型的性能，实验证明该模型能同时得到较高的准确率和召回率。

关键词: 社交网络, 朋友关系, 主题模型, PageRank算法

Abstract:

A new hybrid approach by incorporatin gusers’ interests and users’ friendships together to recommend new friends for target users is proposed.A variation of PageRank—Topic_Friend_PageRank(TFPR) is proposed,which can consider user interests and user friends at same time.Firstly,proposed method uses latent Dirichlet allocation (LDA) to model users’ interests,and weighted-PageRank algorithm to model users’ friendship network,and then merge these two factors into TFPR.This hybrid method models users’ interests and users’ friendships at the same time,and wedemonstrate the effectiveness of proposed hybrid model by using some social network datasets.

Key words: social network, friendship, topic model, PageRank algorithm

尚燕敏,张鹏,曹亚男. 融合链接拓扑结构和用户兴趣的朋友推荐方法[J]. 通信学报, 2015, 36(2): 117-125.

Yan-min SHANG,Peng ZHANG,Ya-nan CAO. New interest-sensitive and network-sensitive method for user recommendation[J]. Journal on Communications, 2015, 36(2): 117-125.

图/表 8

图1

图2

图3

图4

图5

表1

表2

表3

参考文献 18

1	NEWMAN M E J . The structure of scientific collaborationnetworks[A]. Proceedings of the National Academy of Sciences[C]. USA, 2001.404-409.
2	LEO K . A new status index derived from sociometricanalysis[J]. Psychometrika, 1953,18(1):39-43.
3	GLEN J , JENNIFER W . SimRank:a measure of structural-context similarity[A]. Proceedings of the ACMSIGKDD International Conference on Knowledge Discoveryand Data Mining[C]. New Orleans,LA,USA, 2002.538-543.
4	LADA A A , EYTAN A . Friends and neighbors on the web[J]. Social Networks, 2003,25(3):211-230.
5	JORN D , HOLGER E, . STEFAN B.Emergence of a small world from local interactions:modeling acquaintance networks[J]. Physical Review Letters, 2002,88:128701.
6	EMILY M J , MICHELLE G , NEWMAN M E J . The structure of growing social networks[J]. Physical Review Letters E, 2001,64:046132.
7	NEWMAN M E J . Clustering and preferential attachment in growing networks[J]. Physical Review Letters E, 2001,64:025102.
8	GERARD S , MICHAEL J . Introduction to Modern Information Retrieval[M]. McGraw-Hill, 1983.
9	GRANOVETTER M . The strength of weak ties:a network theoryrevisited[J]. Sociological Theory, 1983,1:201-233.
10	MARCO P , SIVA G . Investigating topic models for social media user recommendation[A]. WWW2011[C]. Hyderabad,India, 2011.101-102.
11	KRITI P , JACOB E , SHAY C , et al. Social links from latent topics in microblogs[A]. Proceedings of the NAACL HLT 2010 Workshop onComputational Linguistics in a World of Social Media[C]. Los Angeles,California, 2010.19-20.
12	YIN Z J , MANISH G , TIM W et al. LINKREC:A unified framework for link recommendation with user attributes and graph structure[A]. WWW 2010[C]. North Carolina,USA, 2000.1211-1212.
13	ROHIT P , DOINA C . Predicting friendship linksin social networks using a topic modeling approach[A]. PAKDD2011[C]. 2011.75-86.
14	BLEI D , NG Y A , JORDAN I M . Latent Dirichlet allocation[J]. Journal of Machine Learning Research 3, 2003:993-1022.
15	DANIELE Q , HARRY A , JON C . TweetLDA:supervised topic classification and link prediction in Twitter[A]. WebSci2012[C]. Evaston,Illinois USA, 2012.22-24.
16	LAWRENCE P . The PageRank citation ranking:bring order to Web[J]. Stanford Inforlob, 1998.
17	LARS B , JURE L . Supervised random walks:predicting and recommending links in social networks[A]. WSDM’11[C]. Hong Kong,China, 2011.9-12.
18	TAHER H , HAVELIWALA . Topic sensitive PageRank[A]. WWW 2002[C]. Hawaii,USA, 2011.7-11.

主题数目	迭代次数	预测准确率	召回率	Fmeasure
	100	0.690	0.813	0.746
20	200	0.697	0.835	0.760
	500	0.724	0.833	0.775
	100	0.785	0.843	0.813
50	200	0.810	0.849	0.829
	500	0.821	0.857	0.839
	100	0.834	0.855	0.844
100	200	0.842	0.859	0.850
	500	0.847	0.863	0.855
	100	0.820	0.851	0.835
200	200	0.831	0.854	0.842
	500	0.828	0.848	0.838

迭代次数	向量距离度量方法	Fmeasure
	Cosine	0.844
100
	KL	0.787
	Cosine	0.850
200
	KL	0.798
	Cosine	0.855
500
	KL	0.796

方法	预测准确率	召回率	Fmeasure
LDA-based method^[9]	0.642	0.870	0.739
Graph-follower method	0.902	0.54	0.690
本文模型r=0,d=0.5	0.843	0.860	0.851
本文模型r=1,d=0.5	0.839	0.855	0.850
本文模型r=0.5,d=0.5	0.847	0.863	0.855

[1]	罗智勇, 张玉, 王青, 宋伟伟. 基于贝叶斯攻击图的SDN入侵意图识别算法的研究[J]. 通信学报, 2023, 44(4): 216-225.
[2]	王璿, 张瑜, 周军锋, 陈子阳. 基于社交网络的影响力最大化算法[J]. 通信学报, 2022, 43(8): 151-163.
[3]	彭滔, 钟文韬, 王国军, 罗恩韬, 熊金波, 刘忆宁, Hao Wang. 移动社交网络中面向隐私保护的精确好友匹配[J]. 通信学报, 2022, 43(11): 90-103.
[4]	胡强, 沈嘉吉, 荆广辉, 杜军威. 基于描述语境特征词与改进GSDMM模型的服务聚类方法[J]. 通信学报, 2021, 42(8): 176-187.
[5]	刘真,王娜娜,王晓东,孙永奇. 位置社交网络中谱嵌入增强的兴趣点推荐算法[J]. 通信学报, 2020, 41(3): 197-206.
[6]	顾秋阳, 琚春华, 吴功兴. 基于子图演化与改进蚁群优化算法的社交网络链路预测方法[J]. 通信学报, 2020, 41(12): 21-35.
[7]	陈晶,祁子怡. 基于时序关系的社交网络影响最大化算法研究[J]. 通信学报, 2020, 41(10): 211-221.
[8]	毕俊蕾,李致远. 机会社交网络中基于时变兴趣社区的查询消息路由算法[J]. 通信学报, 2019, 40(9): 86-94.
[9]	李凤华,孙哲,牛犇,曹进,李晖. 跨社交网络的隐私图片分享框架[J]. 通信学报, 2019, 40(7): 1-13.
[10]	黄海平,王凯,汤雄,张东军. 基于边介数模型的差分隐私保护方案[J]. 通信学报, 2019, 40(5): 88-97.
[11]	王晓萌, 方滨兴, 张宏莉, 王星. TSL：基于连接强度的Facebook消息流行度预测模型[J]. 通信学报, 2019, 40(10): 1-9.
[12]	闫光辉, 张萌, 罗浩, 李世魁, 刘婷. 融合高阶信息的社交网络重要节点识别算法[J]. 通信学报, 2019, 40(10): 109-118.
[13]	付钰, 俞艺涵, 吴晓平. 大数据环境下差分隐私保护技术及应用[J]. 通信学报, 2019, 40(10): 157-168.
[14]	邓璐,贾焰,方滨兴,周斌,张涛,刘心. 分布式环境下话题发现算法性能分析[J]. 通信学报, 2018, 39(8): 176-184.
[15]	朱敬华,明骞. LBSN中融合信任与不信任关系的兴趣点推荐[J]. 通信学报, 2018, 39(7): 157-165.

融合链接拓扑结构和用户兴趣的朋友推荐方法

New interest-sensitive and network-sensitive method for user recommendation

在线阅读

PDF下载

可视化

摘要/Abstract

引用本文

使用本文

图/表 8

参考文献 18

相关文章 15

Metrics

推荐阅读 0