An Automatic Label Generation Method for Clustering of Network User Behavior

doi:10.19678/j.issn.1000-3428.0058973

Abstract

Abstract: Existing user behavior clustering methods require the determined size of user behavior data,and the generated cluster labels lack explicit semantics.To solve these problems,this paper proposes an automatic cluster label generation method for clustering analysis of network user behavior.The method applies the Latent Factor Model(LFM) and matrix decomposition method to the raw data of network user behavior for missing value processing.Based on the attribute features of user behavior data,the user behavior cluster is performed and behavior features are added during clustering.At the same time,cluster labels are generated based on behavior feature information to improve the accuracy of user behavior clustering.Experimental results on datasets of Last.fm,Movielens and CiteULike show that the proposed method does not require the determined size of user behavior data,and can automatically generate cluster labels with more explicit semantics while keeping a high clustering accuracy.

Key words: user behavior, clustering, Latent Factor Model(LFM), matrix decomposition, label

摘要： 针对目前多数聚类算法需要事先确定网络用户行为数据规模以及生成的簇标签缺乏明确语义的问题，提出一种用于网络用户行为聚类分析的簇标签自动生成方法。应用潜在因子模型和矩阵分解方法对原始网络用户行为数据补充缺失值，根据网络用户行为数据的属性特征进行用户行为聚类并在聚类过程中增加行为特征，同时利用行为特征信息产生簇标签以提高网络用户行为的聚类准确性。在Last.fm、Movielens和CiteULike数据集上的实验结果表明，该方法无需事先确定网络用户行为数据规模，并且可在保证较高聚类准确率的前提下自动生成语义更明确的簇标签。

关键词: 用户行为, 聚类, 潜在因子模型, 矩阵分解, 标签

CLC Number:

TP309

BI Meng, SHAO Zhong, XU Jian. An Automatic Label Generation Method for Clustering of Network User Behavior[J]. Computer Engineering, 2020, 46(10): 81-87.

毕猛, 邵中, 徐剑. 一种用于网络用户行为聚类的标签自动生成方法[J]. 计算机工程, 2020, 46(10): 81-87.

/ / Recommend / Download Citations

URL: http://www.ecice06.com/EN/10.19678/j.issn.1000-3428.0058973

http://www.ecice06.com/EN/Y2020/V46/I10/81

Figures/Tables 8

References

[1] YANG Jie,QIAO Yuanyuan,ZHANG Xinyu,et al.Characterizing user behavior in mobile Internet[J].IEEE Transactions on Emerging Topics in Computing,2015,3(1):95-106.
[2] REN Xingyi,SONG Meina,SONG Junde.Point-of-Interest recommendation based on the user check-in behavior[J].Chinese Journal of Computers,2017,40(1):28-51.(in Chinese)任星怡,宋美娜,宋俊德.基于用户签到行为的兴趣点推荐[J].计算机学报,2017,40(1):28-51.
[3] SUN Lifeng,WANG Xiaoyan,WANG Zhi,et al.Social-aware video recommendation for online social groups[J].IEEE Transactions on Multimedia,2017,19(3):609-618.
[4] KANG Guosheng,LIU Jianxun,TANG Mingdong,et al.An effective Web service ranking method via exploring user behavior[J].IEEE Transactions on Network and Service Management,2015,12(4):554-564.
[5] BERNASCHINA C,BRAMBILLA M,MAURI A,et al.A big data analysis framework for model-based Web user behavior analytics[C]//Proceedings of International Conference on Web Engineering.Berlin,Germany:Springer,2017:98-114.
[6] WAN Ming,SHANG Wenli,ZENG Peng.Double behavior characteristics for one-class classification anomaly detection in networked control systems[J].IEEE Transactions on Information Forensics and Security,2017,12(12):3011-3023.
[7] SU Qiubin,JIA Zhihao,LU Lu.Research on user behavior clustering algorithm based on mobile application[J].Journal of Intelligent & Fuzzy Systems,2018,35(2):1291-1300.
[8] JOSHI D J,SUPEKAR N,CHAUHAN R,et al.Modeling and detecting change in user behavior through his social media posting using cluster analysis[C]//Proceedings of the 4th ACM IKDD Conference on Data Sciences.New York,USA:ACM Press,2017:1-10.
[9] ZHANG Liping,DENG Song,LI Shiyue.Analysis of power consumer behavior based on the complementation of K-means and DBSCAN[C]//Proceedings of 2017 IEEE Conference on Energy Internet and Energy System Integration.Washington D.C.,USA:IEEE Press,2017:26-28.
[10] ZHANG Jing,DUAN Fu.Improved K-means algorithm with meliorated initial centers[J].Computer Engineering and Design,2013,34(5):1691-1694.(in Chinese)张靖,段富.优化初始聚类中心的改进K-means算法[J].计算机工程与设计,2013,34(5):1691-1694.
[11] ZHOU Zhihua.Machine learning[J].Beijing:Tsinghua University Press,2016.(in Chinese)周志华.机器学习[M].北京:清华大学出版社,2016.
[12] KOREN Y.Collaborative filtering with temporal dynamics[J].Communications of the ACM,2010,53(4):89-97.
[13] GUAN R C,SHI X H,MARCHESE M,et al.Text clustering with seeds affinity propagation[J].IEEE Transactions on Knowledge and Data Engineering,2011,23(4):627-637.
[14] BRIAN K.Revisiting k-means:new algorithms via Bayesian nonparametrics[EB/OL].[2020-06-17].https://arxiv.org/abs/1111.0352.
[15] FARAJIAN M A,MOHAMMADI S.Mining the banking customer behavior using clustering and association rules methods[J].International Journal of Industrial Engineering and Production Research,2010,21(4):239-245.
[16] TREERATPITUK P,CALLAN J.Automatically labeling hierarchical clusters[C]//Proceedings of 2006 Interna-tional Conference on Digital Government Research.New York,USA:ACM Press,2006:167-176.
[17] LÜ Haiyan,ZHANG Jie,WANG Lina.Automatic generation of micro-blog user tags based on clustering analysis[J].Electronic Design Engineering,2015,23(7):67-69,73.(in Chinese)吕海燕,张杰,王丽娜.基于聚类分析的微博用户标签自动生成[J].电子设计工程,2015,23(7):67-69,73.
[18] ZHONG Qing.Research on mobile Internet user behavior classification based on preference tag[J].Mobile Communications,2016(9):93-96.(in Chinese)钟庆.基于喜好标签的移动互联网用户行为分类研究[J].移动通信,2016(9):93-96.
[19] MOHAN A J.An approach to improving cluster labeling and evaluation[D].Twin Cities,USA:University of Minnesota,2014.
[20] LI Huina.A fast and stable cluster labeling method for support vector clustering[J].Journal of Computers,2013,8(12):3251-3256.

Please choose a citation manager

Content to export