基于社会网络面向个性化需求的可信服务推荐

doi:10.3969/j.issn.1000-436X.2013.12.006

摘要/Abstract

摘要：

在当前服务计算和社会计算背景下，针对难以获取满足用户个性化需求的可信 Web 服务问题，给出基于社会网络面向个性化需求的可信 Web 服务推荐模型；设计用户个性化功能需求分解与匹配算法，并利用语义词典提高功能需求语义匹配的准确性；基于个性化功能需求、社会网络节点信任度及服务信任度，设计了一种满足用户个性化需求的可信服务推荐算法，通过对社会网络节点之间、节点与服务之间的信任相关性进行分析，提高服务协同可信推荐性能。算法分析及实验结果表明该方法是有效和可行的。

关键词: 社会网络, 个性化需求, 信任度, 可信服务推荐

Abstract:

In current background of service computing and society computing,it is difficult to obtain the trustworthy Web services to meet users' personalized requirements.To olve the problem,the trustworthy services recommendation model was proposed which orients personalized requirements based on socia twork.The decomposing and matching algorithm was proposed for users' personalized requirements,and the algorithm utilizes the semantic dictionary to improve the se-mantic match of users' requirements.Based on personal zed functional requirements and social trust values of network nodes and the services' direct trust values and indirect trust values,a trustworthy service recommendation algorithm was designed which meets the personalized requirements.By analyzing the correlation among the social network and services' trust values,the method can improve the performance of trustworthy services collaborative recommendation.The analysis of algorithms and experiments' results show that the approach is effective and feasible.

Key words: social network, personalized requirement, trust values, trustworthy services recommendation

张佩云,陈恩红,黄波. 基于社会网络面向个性化需求的可信服务推荐[J]. 通信学报, 2013, 34(12): 49-59.

Pei-yun ZHANG,En-hong CHEN,Bo HUANG. Personalized requirements oriented trustworthy services recommendation based on social network[J]. Journal on Communications, 2013, 34(12): 49-59.

图/表 13

图1

图2

表1

图3

图4

图5

图6

图7

图8

表2

表3

图9

图10

参考文献 20

[1]	韩燕波，陈俊亮，王千祥 . 云计算和服务计算[J]. 计算机学报， 2011,34(12):2251-2252. HAN Y B , CHEN J L , WANG Q X . The preface of special issue of clouding and services computing[J]. Chinese Journal of Computers, 2011,34(12):2251-2252.
[2]	王立才，孟祥武，张玉洁 . 上下文感知推荐系统[J]. 软件学报， 2012,23(1):1-20. WANG L C , MENG X W , ZHANG Y J . Context-aware recommender systems[J]. Journal of Software, 2012,23(1):1-20.
[3]	陈克寒，韩盼盼，吴健 . 基于用户聚类的异构社交网络推荐算法[J]. 计算机学报， 2013,36(2):249-359. CHEN K H , HAN P P , WU J . User clustering based social network recommendation[J]. Chinese Journal of Computers, 2013,36(2):249-359.
[4]	BOBADILLA J , ORTEGA F , HERNANDO A . A collaborative filter-ing similarity measure based on singularities[J]. Information Processing and Management, 2012,(48):204-217.
[5]	WALTER C N , MARIA L H A , RAFAE V G , et al. Social know-ledge-based recommender system application to the movies domain[J]. Expert Systems with Applications, 2012,(39):10990-11000.
[6]	TANG M , JIANG Y , LIU J , et al. Location-aware collaborative filter-ing for QoS-based service recommendation[A]. Proc of IEEE 19th In-ternational Conference on Web Services (ICWS)[C]. Washington DC,USA, 2012.202-209.
[7]	GOLBECK J , HENDLER J . Inferring trust relationships in Web based social networks[J]. ACM Transactions on Internet Technology, 2006,6(4):497-529.
[8]	LIU G , WANG Y , ORGUN M A , et al. Discovering trust networks for the selection of trustworthy service providers in comp x contextual social networks[A]. Proc of IEEE 19th International Conference on Web Services (ICWS),IEEE Computer Society[C]. Washington DC,USA, 2012.384-391.
[9]	LIU G , WANG Y , ORGUN M A . Finding k optimal social trust paths for the selection of trustworthy service providers in mplex social networks[A]. Proc of IEEE 18th International Conference on Web Services (ICWS) IEEE Computer Society[C]. Washington DC,USA, 2011.41-48.
[10]	YAO J , TAN W , NEPAL S , et al. ReputationNet:a reputation engine to enhance servicemap by recommending trusted services[A]. Proc of IEEE Ninth International Conference on Services Computing (SCC),IEEE Computer Society[C]. Washington DC,USA, 2012.454-461.
[11]	ZHONG Y , ZHAO W , YANG J . Personal-hosting RESTful web ser-vices for social network based recommendation[A]. Proc of the 9th International Conference on Service-Oriented Computing(ICSOC)[C]. Paphos,Cyprus, 2011.661-668.
[12]	JAMALI M , ESTER M . A matrix factorization technique with trust propagation for recommendation in social networks[A]. Proc of ACM RecSys,ACM[C]. USA, 2010.135-142.
[13]	MASSA P , AVESANI P . Trust-aware recommender systems[A]. Proc of ACM Recommender Systems Conference[C]. Minneapolis,Minnesota,USA, 2007.17-24.
[14]	王玉祥，乔秀全，李晓峰等. 上下文感知的移动社交网络服务选择机制研究[J]. 计算机学报， 2010,33(11):2126-2135. WANG Y X , QIAO X Q , ,LI X F , et al. Research on context- aware-ness mobile SNS Service selection mechanism[J]. Chinese Journal of Computers, 2010,33(11):2126-2135.
[15]	MA H , KING I , LYU M R . Learning to recommend with explicit and implicit social relations[J]. ACM Transactions on Intelligent Systems and Technology, 2011,2(3):1-19.
[16]	RICHARDSON M , AGRAWAL R , DOMINGOS P . Trust management for the semantic Web[A]. Proc of 2nd International Semantic Web Conference (ISWC),IEEE Computer Society[C]. White Plains,NY,USA, 2003.351-368.
[17]	HANG C , SINAGH M P . Trustworthy service selection and composition[J]. ACM Transactions on Autonomous and Adaptive Systems, 2011,6(1):1-23.
[18]	HURLEY N , ZHANG M . Novelty and diversity in top-N recommen-dation analysis and evaluation[J]. ACM Transactions on Internet Technology, 2011,10(4):1-30.
[19]	韩毅，方滨兴，贾焰等. 基于密度估计的社会网络特征簇挖掘方法[J]. 通信学报， 2012,33(5):38-48. HAN Y , FANG B X , JIA Y , et al. Mining characteristic clusters:a density estimation approach[J]. Journal on Communications, 2012,33(5):38-48.
[20]	胡玉鹏，罗昊，林亚平等. 社会网络中时空周期行为模式挖掘算法[J]. 通信学报， 2013,34(1):8-18. HU Y P , LUO H , LIN Y P , et al. Spatio-temporal periodic behavior mining algorithm for social networks[J]. Journal on Co munications, 2013,34(1):8-18.

参数名称	取值	说明
\|nodes\|	75 879	Epinions数据集中社会网络节点数
\|edges\|	508 837	Epinions数据集中社会网络节点之间的边数
\|S\|	10 000	社会网络拥有的服务数
\|NList\|	9	社会网络的节点的平均邻居数
threshold	0.5	算法2的个性化需求匹配阈值
trustDegree	0.5	算法2的个性化信任度阈值

算法	社会网络节点数					平均偏差
算法	100	300	700	900	1 100	平均偏差
DTR	0.014 8	0.067 1	0.171 5	0.048 8	0.040 8	0.068 6
ITR	0.114 5	0.016 5	0.014 5	0.070 9	0.074 5	0.058 18
PTR	0.257	0.177	0.157	0.216	0.021	0.165 6
NTR	0.034 4	0.081 9	0.187 4	0.002 3	0.137 6	0.088 72

算法	社会网络节点数					平均偏差
算法	100	300	700	900	1 100	平均偏差
DTR	0.157	0.005 6	0.067 1	0.054 4	0.038 4	0.064 5
ITR	0.209 5	0.016 9	0.088 5	0.098 8	0.039 1	0.090 6
PTR	0.175 1	0.057 1	0.116 9	0.094 5	0.020 9	0.092 9
NTR	0.134 5	0.058 5	0.095 1	0.005 8	0.024 9	0.063 8