利用Plackett-Luce模型的在线服务评价方法
|
摘要
每一位用户在选择服务过程都中具有自己独特的标准,这就导致不同用户对同一服务的评价不具备可比较性.通过不具备可比较性的服务评价进行简单计算得到的评价结果会影响用户选择.为此,本文提出一种利用Plackett-Luce模型的在线服务评价方法.首先根据用户服务评分计算获取在线服务的偏好关系,并根据偏好关系得到占优次数;其次将服务占优次数转化为服务排序权重,并建立Plackett-Luce模型的迭代函数,对排序权重值进行迭代计算;最后把排序权重值转化的概率作为服务评价结果.通过基于公开数据集的实验验证了本文所提出方法的合理性和有效性.
Each user has their own unique criteria in the selection of the service process,which results in different users' evaluation of the same service is not comparable. The evaluation results obtained by simple calculations without comparable service evaluations are difficult to reflect the pros and cons of the services. To this end,this paper proposes an online service evaluation method based on the Plackett-Luce model. Firstly,according to the user service score,the user's preference relationship with the online service is obtained,and the optimal number of times is obtained according to the preference relationship. Secondly,the service dominant time is converted into the service ranking weight,and the iterative function of the Plackett-Luce model is established,and the sorting weight value is iterated. The calculation is performed; finally,the probability of the conversion weight value is converted as the service evaluation result.The rationality and effectiveness of the proposed method are verified by experiments based on public data sets.
Each user has their own unique criteria in the selection of the service process,which results in different users' evaluation of the same service is not comparable. The evaluation results obtained by simple calculations without comparable service evaluations are difficult to reflect the pros and cons of the services. To this end,this paper proposes an online service evaluation method based on the Plackett-Luce model. Firstly,according to the user service score,the user's preference relationship with the online service is obtained,and the optimal number of times is obtained according to the preference relationship. Secondly,the service dominant time is converted into the service ranking weight,and the iterative function of the Plackett-Luce model is established,and the sorting weight value is iterated. The calculation is performed; finally,the probability of the conversion weight value is converted as the service evaluation result.The rationality and effectiveness of the proposed method are verified by experiments based on public data sets.
引文
[1]J?sang A,Ismail R,Boyd C.A survey of trust and reputation systems for online service provision[J].Decision Support Systems,2007,43(2):618-644.
[2]Zhang Wei,Liu Lu,Zhu Yan-chun,et al.Recent developments and prospect of online reputation system[J].Control and Dceision,2005,20(11):1201-1207.
[3]Golbeck J.Weaving a web of turst[J].Science,2008,321(5869):1640-1641.
[4]Yao Yuan,Sini Ruohomaa,Xu Feng,et al.Addressing common vulnerabilities of reputation systems for electronic commerce[J].Journal of Theoretical and Applied Electronic Commerce Research,2012,7(1):1-20.
[5]Fu Xiao-Dong,Yue Kun,Liu Li,et al.Aggregating ordinal user preferences for effective reputation computation of online services[C]//IEEE International Conference on Web Services,2016:554-561.
[6]Schall D,Skopik F,Dustdar S.Expert discovery and Interaction in mixed service-oriented system[J].IEEE Transaction on Services Compution,2015,5(2):233-245.
[7]Allahbakhsh M,Ignjatovic A,Motahari-Nezhad H R,et al.Robust evaluation of products and review ers in social rating system[J].World Wide Web,2014,18(1):73-109.
[8]Fu Xiao-dong,Zou Ping,Jiang Ying,et al.Web service reputation measurement based on quality of service similarity[J].Computer Integrated M anufacturing Systems,2008,14(3):619-624.
[9]Luce,Robert Duncan.Individual choice behavior:a theoretical analysis[J].Journal of the American Statistical Association,2005,67(1):1-15.
[10]Plackett,Robin L.The analysis of permutations[J].Journal of the Royal Statistical Society,Series C(Applied Statistics),1975,24(2):193-202.
[11]David H A.The method of paired comparisons[J].Biometrics,1963,152(2):1-21.
[12]Abdu-Muhmin A G.Contingent decision behavior;effect of number of alternative to be selectef on consumers'decision processes[J].Journal of Consumer Psychology,1999,8(1):91-111.
[13]Tian Pei-liang,Liu Yuan-chao,Liu Ming,et a1.Research of product ranking technology based on opinion mining[C]//Intelligent Computation Technology,and Automation(ICICTA'09),Second International Conference on IEEE,2009,4:239-243.
[14]Najmi E,Hashmi K,Malik Z,et a1.CAPRA:a comprehensive approach to product ranking using customer review s[J].Computing,2015,97(8):843-867.
[15]Gangothri V,Saranya S,Venkataraman D.Engender product ranking and recommendation using customer feedback[C]//Proceedings of the International Conference on Soft Computing Systems,Springer India,2016:851-859.
[16]Zhang Xiu-zhen,Cui Li-shan,Wang Yan.Computing multi-dimensional trust by mining e-commerce comments[J].IEEE Transactions on Know ledge&Data Engineering,2013,26(13):1631-1643.
[17]Fan Yue-qin,Zhang Yong-sheng,Chang Liang.Research on the trust reasoning of multi dimension atribute oriented personalized service[J].Journal of Chineses Computer Systems,2016,37(9):1986-1990.
[18]Sun Ruo-nan,Zhang Bin,Liu Ting-ting.Service ranking method based on improved entropy TOPSIS[J].Journal of Chineses Computer Systems,2017,38(6):1221-1226.
[19]Chen Ying,Fu Xiao-dong,Yue Kun,et-al.An online products evaluation methon considering the inconsistency of user rating criteria[J].Journal of Chineses Computer Systems,2017,38(6):1317-1322.
[20]Yin Yan,Fu Xiao-dong,Liu Li,et al.Group evaluation method for online commodities using Copeland social choice theory[J].Journal of Chineses Computer Systems,2018,39(6):1201-1207.
[21]Okasha S.Theory choice and social choice:Kuhn versus arrow[J].Mind,2011,120(477):83-115.
[22]Hunter,David R.MM algorithms for generalized bradley-terry models[J].The Annals of Statistics,2004,32:384-406.
[23]Li Li-sha,Guo Fen-zhuo,Qin Su-juan.A listwise collaborative filtering based on Plackett-Luce model[C]//3rd IEEE International Conference on Computer and Communications(ICCC),2017:2159-2163.
[24]Fu Xiao-dong,Li Jun,Liu Li,et al.Evaluating online services based on a ranked pairs social choice function[J].Journal of Tsinghua U-niversity(Sicence and Technology),2018,58(8):715-724.
[25]Harper F M,Konstan J A.The movieLens datasets:history and context[J].ACM Transactions on Interactive Intelligent Systems,2015,5(4),https://doi.org/10.1145/2827872.
[26]Herings J J,Houba H.The Condorcet paradox revisited[J].Social Choice and Welfare,2016,47(1):141-186.
[2]张巍,刘鲁,朱艳春,等.在线信誉系统研究现状与展望[J].控制与决策,2005,20(11):1201-1207.
[8]付晓东,邹平,姜瑛,等.基于质量相似度的Web服务信誉度量[J].计算机集成制造系统,2008,14(3):619-624.
[17]樊悦芹,张永胜,常良,面向个性化服务的多维属性信任推理研究[J].小型微型计算机系统,2016,37(9):1986-1990.
[18]孙若男,张斌,刘婷婷.一种使用改进熵权TOPSIS的Web服务质量评价排序方法[J].小型微型计算机系统,2017,38(6):1221-1226.
[19]陈颖,付晓东,岳昆,等.一种考虑用户评分标准不一致性的在线商品评价方法[J].小型微型计算机系统,2017,38(6):1317-1322.
[20]殷岩,付晓东,刘骊,等.利用Copeland社会选择理论的在线商品群体评价[J].小型微型计算机系统2018,39(6):1201-1207.
[24]付晓东,李俊,刘骊,等.基于排序对社会选择函数的在线服务评价[J].清华大学学报(自然科学版),2018,58(8):715-724.
1 https://www. amazon. cn/
2 https://www. tmall. com/
3 https://www. ebay. cn/
4 https://www. taobao. com/
[2]Zhang Wei,Liu Lu,Zhu Yan-chun,et al.Recent developments and prospect of online reputation system[J].Control and Dceision,2005,20(11):1201-1207.
[3]Golbeck J.Weaving a web of turst[J].Science,2008,321(5869):1640-1641.
[4]Yao Yuan,Sini Ruohomaa,Xu Feng,et al.Addressing common vulnerabilities of reputation systems for electronic commerce[J].Journal of Theoretical and Applied Electronic Commerce Research,2012,7(1):1-20.
[5]Fu Xiao-Dong,Yue Kun,Liu Li,et al.Aggregating ordinal user preferences for effective reputation computation of online services[C]//IEEE International Conference on Web Services,2016:554-561.
[6]Schall D,Skopik F,Dustdar S.Expert discovery and Interaction in mixed service-oriented system[J].IEEE Transaction on Services Compution,2015,5(2):233-245.
[7]Allahbakhsh M,Ignjatovic A,Motahari-Nezhad H R,et al.Robust evaluation of products and review ers in social rating system[J].World Wide Web,2014,18(1):73-109.
[8]Fu Xiao-dong,Zou Ping,Jiang Ying,et al.Web service reputation measurement based on quality of service similarity[J].Computer Integrated M anufacturing Systems,2008,14(3):619-624.
[9]Luce,Robert Duncan.Individual choice behavior:a theoretical analysis[J].Journal of the American Statistical Association,2005,67(1):1-15.
[10]Plackett,Robin L.The analysis of permutations[J].Journal of the Royal Statistical Society,Series C(Applied Statistics),1975,24(2):193-202.
[11]David H A.The method of paired comparisons[J].Biometrics,1963,152(2):1-21.
[12]Abdu-Muhmin A G.Contingent decision behavior;effect of number of alternative to be selectef on consumers'decision processes[J].Journal of Consumer Psychology,1999,8(1):91-111.
[13]Tian Pei-liang,Liu Yuan-chao,Liu Ming,et a1.Research of product ranking technology based on opinion mining[C]//Intelligent Computation Technology,and Automation(ICICTA'09),Second International Conference on IEEE,2009,4:239-243.
[14]Najmi E,Hashmi K,Malik Z,et a1.CAPRA:a comprehensive approach to product ranking using customer review s[J].Computing,2015,97(8):843-867.
[15]Gangothri V,Saranya S,Venkataraman D.Engender product ranking and recommendation using customer feedback[C]//Proceedings of the International Conference on Soft Computing Systems,Springer India,2016:851-859.
[16]Zhang Xiu-zhen,Cui Li-shan,Wang Yan.Computing multi-dimensional trust by mining e-commerce comments[J].IEEE Transactions on Know ledge&Data Engineering,2013,26(13):1631-1643.
[17]Fan Yue-qin,Zhang Yong-sheng,Chang Liang.Research on the trust reasoning of multi dimension atribute oriented personalized service[J].Journal of Chineses Computer Systems,2016,37(9):1986-1990.
[18]Sun Ruo-nan,Zhang Bin,Liu Ting-ting.Service ranking method based on improved entropy TOPSIS[J].Journal of Chineses Computer Systems,2017,38(6):1221-1226.
[19]Chen Ying,Fu Xiao-dong,Yue Kun,et-al.An online products evaluation methon considering the inconsistency of user rating criteria[J].Journal of Chineses Computer Systems,2017,38(6):1317-1322.
[20]Yin Yan,Fu Xiao-dong,Liu Li,et al.Group evaluation method for online commodities using Copeland social choice theory[J].Journal of Chineses Computer Systems,2018,39(6):1201-1207.
[21]Okasha S.Theory choice and social choice:Kuhn versus arrow[J].Mind,2011,120(477):83-115.
[22]Hunter,David R.MM algorithms for generalized bradley-terry models[J].The Annals of Statistics,2004,32:384-406.
[23]Li Li-sha,Guo Fen-zhuo,Qin Su-juan.A listwise collaborative filtering based on Plackett-Luce model[C]//3rd IEEE International Conference on Computer and Communications(ICCC),2017:2159-2163.
[24]Fu Xiao-dong,Li Jun,Liu Li,et al.Evaluating online services based on a ranked pairs social choice function[J].Journal of Tsinghua U-niversity(Sicence and Technology),2018,58(8):715-724.
[25]Harper F M,Konstan J A.The movieLens datasets:history and context[J].ACM Transactions on Interactive Intelligent Systems,2015,5(4),https://doi.org/10.1145/2827872.
[26]Herings J J,Houba H.The Condorcet paradox revisited[J].Social Choice and Welfare,2016,47(1):141-186.
[2]张巍,刘鲁,朱艳春,等.在线信誉系统研究现状与展望[J].控制与决策,2005,20(11):1201-1207.
[8]付晓东,邹平,姜瑛,等.基于质量相似度的Web服务信誉度量[J].计算机集成制造系统,2008,14(3):619-624.
[17]樊悦芹,张永胜,常良,面向个性化服务的多维属性信任推理研究[J].小型微型计算机系统,2016,37(9):1986-1990.
[18]孙若男,张斌,刘婷婷.一种使用改进熵权TOPSIS的Web服务质量评价排序方法[J].小型微型计算机系统,2017,38(6):1221-1226.
[19]陈颖,付晓东,岳昆,等.一种考虑用户评分标准不一致性的在线商品评价方法[J].小型微型计算机系统,2017,38(6):1317-1322.
[20]殷岩,付晓东,刘骊,等.利用Copeland社会选择理论的在线商品群体评价[J].小型微型计算机系统2018,39(6):1201-1207.
[24]付晓东,李俊,刘骊,等.基于排序对社会选择函数的在线服务评价[J].清华大学学报(自然科学版),2018,58(8):715-724.
1 https://www. amazon. cn/
2 https://www. tmall. com/
3 https://www. ebay. cn/
4 https://www. taobao. com/