Multi-label classification in network environments is becoming a key area of data mining research as its applications are increasing dramatically.Relational classification models,which predict class labels of linked neighbors according to the ones of the given nodes,have been shown to outperform traditional multi-label classifiers.However,existing relational classification models neither make full use of neighbor information,nor predict the isolated nodes' labels,which are popularly existing in relational networks.In this paper,we present a multi-label relational classifier (MORN) that mines both second-order neighbors for non-isolated nodes and high-order neighbors for isolated nodes.MORN has been conducted on real datasets and it demonstrates that our proposed classifier outperforms existing relational classification models.
[1] S Macskassy,F Provost.A simple relational classifier[A].Proceedings of the Second Workshop on Multi-Relational Data Mining at ACM SIGKDD[C].Washington DC,USA:ACM,2003.64-76.
[2] P SEN,G Namata,M Bilgic,L Getoor,B Gallagher,T Eliassi-Rad.Collective classification in network data[J].AI Magazine,2008,29(3):93-106.
[3] X Wang,G Sukthankar.Multi-label relational neighbor classification using social context features[A].Proceedings of the ACM SIGKDD[C].Chicago,USA:ACM,2013.464-472.
[4] J Neville,D Jensen,L Friedland,M Hay.Learning relational probability trees[A].Proceedings of the ACM SIGKDD[C].Washington DC,USA:ACM,2003.625-630.
[5] B Taskar,P Abbeel,D Koller.Discriminative probabilistic models for relational data[A].Proceedings of the UAI[C].Edmonton,Canada:Morgan Kaufmann,2002.485-492.
[6] M Zhang,K Zhang.Multi-label learning by exploiting label dependency[A].Proceedings of the ACM SIGKDD[C].Washington DC,USA:ACM,2010.999-1007.
[7] C Aggarwal.Social Network Data Analytics[M].Berlin:Springer,2011.115-148.
[8] E Hullermeier,J Furnkranz,W Cheng,K Brinker.Label ranking by learning pairwise preferences[J].Artificial Intelligence,2008,172(16):1897-1916.
[9] M Zhang,Z Zhou.A k-nearest neighbor based algorithm for multi-label classification[A].Proceedings of the IEEE International Conference on Granular Computing[C].Beijing,China:IEEE,2005.718-721.
[10] 郑伟,王朝坤,刘璋,王建民.一种基于随机游走模型的多标签分类算法[J].计算机学报,2010,33(8):1418-1426. W Zheng,C Wang,Z Liu,J Wang.A multi-label classification algorithm based on random walk model[J].Chinese Journal of Computers,2010,33(8):1418-1426.(in Chinese)
[11] L Tang,H Liu.Scalable learning of collective behavior based on sparse social dimensions[A].Proceedings of the ACM CIKM[C].Hong Kong,China:ACM,2009.1107-1116.
[12] J Neville,D Jensen.Iterative classification in relational data[A].Proceedings of the AAAI Workshop on Learning Statistical Models from Relational Data[C].Austin,USA:AAAI,2000.42-49.
[13] S Macskassy,F Provost.Classification in networked data:a toolkit and a univariate case study[J].Journal of Machine Learning,2007,8(5):935-983.
[14] M Newman.Networks:An Introduction[M].Oxford:Oxford University Press,2010.
[15] M McPherson,L Smith-lovin,J Cook.Birds of a feather:Homophily in social networks[J].Annual Review of Sociology,2001,27(1):415-444.
[16] J Neville,B Gallagher,T Eliassi-Rad,T Wang,Correcting evaluation bias of relational classifiers with network cross validation[J].Knowledge and Information Systems,2012,30(1):31-55.
2016, Vol. 44 
