Class-aware tensor factorization for multi-relational classification

Abstract In this paper, we propose a tensor factorization method, called CLASS-RESCAL, which associates the class labels of data samples with their latent representations. Specifically, we extend RESCAL to produce a semi-supervised factorization method that combines a classification error term with the standard factor optimization process. CLASS-RESCAL assimilates information from all the relations of the tensor, while also taking into account classification performance. This procedure forces the data samples within the same class to have similar latent representations. Experimental results on several real-world social network data indicate this is a promising approach for multi-relational classification tasks.

[1]  Hans-Peter Kriegel,et al.  Factorizing YAGO: scalable machine learning for linked data , 2012, WWW.

[2]  Deren Han,et al.  Nonnegative tensor factorizations using an alternating direction method , 2013 .

[3]  Tamara G. Kolda,et al.  All-at-once Optimization for Coupled Matrix and Tensor Factorizations , 2011, ArXiv.

[4]  David F. Gleich,et al.  Multilinear PageRank , 2014, SIAM J. Matrix Anal. Appl..

[5]  Zhengyang Wang,et al.  Large-Scale Learnable Graph Convolutional Networks , 2018, KDD.

[6]  Max Welling,et al.  Semi-Supervised Classification with Graph Convolutional Networks , 2016, ICLR.

[7]  Jure Leskovec,et al.  {SNAP Datasets}: {Stanford} Large Network Dataset Collection , 2014 .

[8]  Jure Leskovec,et al.  node2vec: Scalable Feature Learning for Networks , 2016, KDD.

[9]  Hans-Peter Kriegel,et al.  A Three-Way Model for Collective Learning on Multi-Relational Data , 2011, ICML.

[10]  Qiang Ma,et al.  Dual Graph Convolutional Networks for Graph-Based Semi-Supervised Classification , 2018, WWW.

[11]  Lars Schmidt-Thieme,et al.  Multi-Relational Learning at Scale with ADMM , 2016, ArXiv.

[12]  Steffen Staab,et al.  TripleRank: Ranking Semantic Web Data by Tensor Decomposition , 2009, SEMWEB.

[13]  Yunming Ye,et al.  MultiRank: co-ranking for objects and relations in multi-relational data , 2011, KDD.

[14]  Max Welling,et al.  Variational Graph Auto-Encoders , 2016, ArXiv.

[15]  M. McPherson,et al.  Birds of a Feather: Homophily in Social Networks , 2001 .

[16]  Max Welling,et al.  Modeling Relational Data with Graph Convolutional Networks , 2017, ESWC.

[17]  Lars Schmidt-Thieme,et al.  Learning optimal ranking with tensor factorization for tag recommendation , 2009, KDD.

[18]  Jimeng Sun,et al.  MetaFac: community discovery via relational hypergraph factorization , 2009, KDD.

[19]  R. Harshman,et al.  PARAFAC: parallel factor analysis , 1994 .

[20]  Ee-Peng Lim,et al.  Structural Analysis in Multi-Relational Social Networks , 2012, SDM.

[21]  T. Kolda.  Multilinear operators for higher-order decompositions , 2006 .

[22]  Rasmus Bro,et al.  Understanding data fusion within the framework of coupled matrix and tensor factorizations , 2013 .

[23]  Vagelis Hristidis,et al.  ObjectRank: a system for authority-based search on databases , 2006, SIGMOD Conference.

[24]  Chen Guolong,et al.  Mining Social Influence in Microblogging via Tensor Factorization Approach , 2013, 2013 International Conference on Cloud Computing and Big Data.

[25]  Jeffrey Dean,et al.  Efficient Estimation of Word Representations in Vector Space , 2013, ICLR.

[26]  L. Tucker,et al.  Some mathematical notes on three-mode factor analysis , 1966, Psychometrika.

[27]  Joan Ferrante-Wallace,et al.  Sociology : a global perspective , 2015 .

[28]  Tamara G. Kolda,et al.  Temporal Analysis of Semantic Graphs Using ASALSAN , 2007, Seventh IEEE International Conference on Data Mining (ICDM 2007).

[29]  Yunming Ye,et al.  HAR: Hub, Authority and Relevance Scores in Multi-Relational Data for Query Search , 2012, SDM.

[30]  Hisashi Kashima,et al.  Tensor factorization using auxiliary information , 2011, Data Mining and Knowledge Discovery.

[31]  Jia Yan,et al.  Mining topical influencers based on the multi-relational network in micro-blogging sites , 2013, China Communications.

[32]  Xu Tan,et al.  Supervised Nonnegative Tensor Factorization with Maximum-Margin Constraint , 2013, AAAI.

[33]  Pauli Miettinen,et al.  Discovering facts with boolean tensor tucker decomposition , 2013, CIKM.

[34]  Jiawei Han,et al.  Mining topic-level influence in heterogeneous networks , 2010, CIKM.

[35]  Joan Ferrante,et al.  Sociology: A Global Perspective , 1992 .

[36]  Wei-Ying Ma,et al.  Object-level ranking: bringing order to Web objects , 2005, WWW '05.

[37]  Myra Spiliopoulou,et al.  A framework for validating the merit of properties that predict the influence of a twitter user , 2015, Expert Syst. Appl..

[38]  Partha P. Talukdar,et al.  Relation Schema Induction using Tensor Factorization with Side Information , 2016, EMNLP.

[39]  Nikos D. Sidiropoulos,et al.  Parallel Algorithms for Constrained Tensor Factorization via Alternating Direction Method of Multipliers , 2014, IEEE Transactions on Signal Processing.

[40]  James R. Foulds,et al.  Collective Spammer Detection in Evolving Multi-Relational Social Networks , 2015, KDD.

[41]  Johan Bollen,et al.  Co-authorship networks in the digital library research community , 2005, Inf. Process. Manag..

[42]  Alioune Ngom,et al.  Non-negative matrix and tensor factorization based classification of clinical microarray gene expression data , 2010, 2010 IEEE International Conference on Bioinformatics and Biomedicine (BIBM).