{"type":"ScholarlyArticle","title":"Discovering Relevance-Dependent Bicluster Structure from Relational Data","authors":[{"name":"Iku Ohama","slug":"iku-ohama","given":"Iku","family":"Ohama","affiliation":"‡ Panasonic Corporation , Japan Graduate School of Information Science and Technology ,"},{"name":"Takuya Kida","slug":"takuya-kida","given":"Takuya","family":"Kida","affiliation":"‡ Panasonic Corporation , Japan Graduate School of Information Science and Technology ,"},{"name":"Hiroki Arimura","slug":"hiroki-arimura","given":"Hiroki","family":"Arimura","affiliation":"‡ Panasonic Corporation , Japan Graduate School of Information Science and Technology ,; Hokkaido University , Japan"}],"abstract":"In this paper, we propose a statistical model for relevance-dependent biclustering to analyze relational data. The proposed model factorizes relational data into bicluster structure with two features: (1) each object in a cluster has a relevance value, which indicates how strongly the object relates to the cluster and (2) all clusters are related to at least one dense block. These features simplify the task of understanding the meaning of each cluster because only a few highly relevant objects need to be inspected. We introduced the Relevance-Dependent Bernoulli Distribution (R-BD) as a prior for relevance-dependent binary matrices and proposed the novel Relevance-Dependent Infinite Biclustering (R-IB) model, which automatically estimates the number of clusters. Posterior inference can be performed efficiently using a collapsed Gibbs sampler because the parameters of the R-IB model can be fully marginalized out. Experimental results show that the R-IB extracts more essential bicluster structure with better computational efficiency than conventional models. We further observed that the biclustering results obtained by R-IB facilitate interpretation of the meaning of each cluster.","doi":"10.24963/ijcai.2017/359","url":"https://ijcai.org/proceedings/2017/359","pdf":"https://ijcai.org/proceedings/2017/0359.pdf","markdown":"https://ijcai.org/proceedings/2017/359.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/359","volume":{"slug":"2017","title":"Proceedings of the Twenty-Sixth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2017","url":"https://ijcai.org/proceedings/2017/","year":2017,"location":"Melbourne, Australia","dates":"19-25 August 2017","editors":"Carles Sierra","isbn":"978-0-9992411-0-3","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2017"},"section":"Machine Learning A-R","track":"Main track","number":359,"pages":{"first":2578,"last":2584},"keywords":["Machine Learning: Learning Graphical Models","Machine Learning: Data Mining","Machine Learning: Unsupervised Learning","Machine Learning: Relational Learning"],"videos":[],"references":[{"position":1,"raw":"[Airoldi et al., 2008] Edoardo M. Airoldi, David M. Blei, Stephen E. Fienberg, and Eric P. Xing. Mixed membership stochastic blockmodels. J. Mach. Learn. Res., 9:1981- 2014, 2008.","title":"Mixed membership stochastic blockmodels","authors":"M Edoardo, David M Airoldi, Stephen E Blei, Eric P Fienberg, Xing","year":"2008","venue":"J. Mach. Learn. Res","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Aldous, 1985] David Aldous. Exchangeability and related topics. In Ecole d'Ete de Probabilities de Saint-Flour XIII, pages 1-198. 1985.","title":"Exchangeability and related topics","authors":"David Aldous","year":"1985","venue":"Ecole d'Ete de Probabilities de Saint-Flour XIII","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Araujo et al., 2014] Miguel Araujo, Stephan Günnemann, Gonzalo Mateos, and Christos Faloutsos. Beyond blocks: Hyperbolic community detection. In Proc. ECML-PKDD, pages 50-65, 2014.","title":"Beyond blocks: Hyperbolic community detection","authors":"Araujo","year":"2014","venue":"Proc. ECML-PKDD","doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Blackwell and MacQueen, 1973] David Blackwell and James B. MacQueen. Ferguson distributions via Polya urn schemes. The Annals of Statistics, 1:353-355, 1973.","title":"Ferguson distributions via Polya urn schemes","authors":"Macqueen Blackwell, David Blackwell, James B Macqueen","year":"1973","venue":"The Annals of Statistics","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Davis and Goadrich, 2006] Jesse Davis and Mark Goad- rich. The relationship between precision-recall and ROC curves. In Proc. ICML, pages 233-240, 2006.","title":"Jesse Davis and Mark Goadrich. The relationship between precision-recall and ROC curves","authors":"Goadrich Davis","year":"2006","venue":"Proc. ICML","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Escobar and West, 1994] Michael D. Escobar and Mike West. Bayesian density estimation and inference using mixtures. J. Am. Stat. Assoc., 90:577-588, 1994.","title":"Bayesian density estimation and inference using mixtures","authors":"D Michael, Mike Escobar, West","year":"1994","venue":"J. Am. Stat. Assoc","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Fu et al., 2009] Wenjie Fu, Le Song, and Eric P. Xing. Dy- namic mixed membership blockmodel for evolving net- works. In Proc. ICML, pages 329-336, 2009.","title":"Dynamic mixed membership blockmodel for evolving networks","authors":"Wenjie Fu, Le Song, Eric P Xing","year":"2009","venue":"Proc. ICML","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Geyer, 2007] Charles J. Geyer. Lower-truncated Pois- son and negative binomial distributions. Technical report, Working Paper Written for the Software R. University of Minnesota, MN (available: http://cran.r- project.org/web/packages/aster/vignettes/trunc.pdf), 2007.","title":"Lower-truncated Poisson and negative binomial distributions","authors":"Charles J Geyer","year":"2007","venue":"Working Paper Written for the Software R","doi":null,"url":"http://cran.r-project.org/web/packages/aster/vignettes/trunc.pdf","internal_url":null},{"position":9,"raw":"[Ho et al., 2011] Qirong Ho, Ankur P. Parikh, Le Song, and Eric P. Xing. Multiscale community blockmodel for net- work exploration. Proc. AISTATS, pages 333-341, 2011.","title":"Multiscale community blockmodel for network exploration","authors":"Qirong Ho, Ankur P Parikh, Le Song, Eric P Xing","year":"2011","venue":"Proc. AISTATS","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Ishiguro et al., 2010] Katsuhiko Ishiguro, Tomoharu Iwata, Naonori Ueda, and Joshua B. Tenenbaum. Dynamic infi- nite relational model for time-varying relational data anal- ysis. In Proc. NIPS, pages 919-927, 2010.","title":"Dynamic infinite relational model for time-varying relational data analysis","authors":"Katsuhiko Ishiguro, Tomoharu Iwata, Naonori Ueda, Joshua B Tenenbaum","year":"2010","venue":"Proc. NIPS","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Ishiguro et al., 2012] Katsuhiko Ishiguro, Naonori Ueda, and Hiroshi Sawada. Subset infinite relational models. In Proc. AISTATS, pages 547-555, 2012.","title":"Subset infinite relational models","authors":"Katsuhiko Ishiguro, Naonori Ueda, Hiroshi Sawada","year":"2012","venue":"Proc. AISTATS","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Ishiguro et al., 2016] Katsuhiko Ishiguro, Issei Sato, Masahiro Nakano, Akisato Kimura, and Naonori Ueda. Infinite plaid models for infinite bi-clustering. In Proc. AAAI, pages 1701-1708, 2016.","title":"Infinite plaid models for infinite bi-clustering","authors":"Katsuhiko Ishiguro, Issei Sato, Masahiro Nakano, Akisato Kimura, Naonori Ueda","year":"2016","venue":"Proc. AAAI","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Kemp et al., 2006] Charles Kemp, Joshua B. Tenenbaum, Thomas L. Griffiths, Takeshi Yamada, and Naonori Ueda. Learning systems of concepts with an infinite relational model. In Proc. AAAI, volume 1, pages 381-388, 2006.","title":"Learning systems of concepts with an infinite relational model","authors":"Charles Kemp, Joshua B Tenenbaum, Thomas L Griffiths, Takeshi Yamada, Naonori Ueda","year":"2006","venue":"Proc. AAAI","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Klimat and Yang, 2004] Bryan Klimat and Yiming Yang. The Enron corpus: A new dataset for email classification research. In Proc. ECML, pages 217-226, 2004.","title":"The Enron corpus: A new dataset for email classification research","authors":"Bryan Klimat, Yiming Yang","year":"2004","venue":"Proc. ECML","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Liu, 1994] Jun S. Liu. The collapsed Gibbs sampler in bayesian computations with applications to a gene regula- tion problem. J. Am. Stat. Assoc., 89(427):958-966, 1994.","title":"The collapsed Gibbs sampler in bayesian computations with applications to a gene regulation problem","authors":"S Jun, Liu","year":"1994","venue":"J. Am. Stat. Assoc","doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Mørup et al., 2011] Morten Mørup, Mikkel N. Schmidt, and Lars Kai Hansen. Infinite multiple membership re- lational modeling for complex networks. In Proc. MLSP, pages 1-6, 2011.","title":"Infinite multiple membership relational modeling for complex networks","authors":"Mørup","year":"2011","venue":"Proc. MLSP","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"MOV, as of 2003] MovieLens dataset, http://www.grouplens.org/, as of 2003.","title":"MovieLens dataset","authors":"Mov","year":"2003","venue":null,"doi":null,"url":"http://www.grouplens.org/,asof2003","internal_url":null},{"position":18,"raw":"[Nakano et al., 2014] Masahiro Nakano, Katsuhiko Ishig- uro, Akisato Kimura, Takeshi Yamada, and Naonori Ueda. Rectangular tiling process. In Proc. ICML, pages 361-369, 2014.","title":"Rectangular tiling process","authors":"Masahiro Nakano, Katsuhiko Ishiguro, Akisato Kimura, Takeshi Yamada, Naonori Ueda","year":"2014","venue":"Proc. ICML","doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Newman et al., 2009] David Newman, Arthur U. Asuncion, Padhraic Smyth, and Max Welling. Distributed algorithms for topic models. J. Mach. Learn. Res., 10:1801-1828, 2009.","title":"Distributed algorithms for topic models","authors":"David Newman, Arthur U Asuncion, Padhraic Smyth, Max Welling","year":"2009","venue":"J. Mach. Learn. Res","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Nowicki and Snijders, 2001] Krzysztof Nowicki and Tom A. B. Snijders. Estimation and prediction for stochastic blockstructures. J. Am. Stat. Assoc., 96(455):1077-1087, 2001.","title":"Estimation and prediction for stochastic blockstructures","authors":"Krzysztof Nowicki, A B Tom, Snijders","year":"2001","venue":"J. Am. Stat. Assoc","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Ohama et al., 2013] Iku Ohama, Hiromi Iida, Takuya Kida, and Hiroki Arimura. An extension of the infinite relational model incorporating interaction between objects. In Proc. PAKDD (2), pages 147-159, 2013.","title":"An extension of the infinite relational model incorporating interaction between objects","authors":"Iku Ohama, Hiromi Iida, Takuya Kida, Hiroki Arimura","year":"2013","venue":"Proc. PAKDD (2)","doi":null,"url":null,"internal_url":null},{"position":22,"raw":"[Ohama et al., 2015] Iku Ohama, Takuya Kida, and Hiroki Arimura. Multi-layered framework for modeling relation- ships between biased objects. In Proc. SDM, pages 819- 827, 2015.","title":"Multi-layered framework for modeling relationships between biased objects","authors":"Iku Ohama, Takuya Kida, Hiroki Arimura","year":"2015","venue":"Proc. SDM","doi":null,"url":null,"internal_url":null},{"position":23,"raw":"[Ohama et al., 2016] Iku Ohama, Hiromi Iida, Takuya Kida, and Hiroki Arimura. The relevance dependent infinite rela- tional model for discovering co-cluster structure from rela- tionships with structured noise. IEICE Trans. on Inf. Sys., E99-D(4):1139-1152, 2016.","title":"The relevance dependent infinite relational model for discovering co-cluster structure from relationships with structured noise","authors":"Iku Ohama, Hiromi Iida, Takuya Kida, Hiroki Arimura","year":"2016","venue":"IEICE Trans. on Inf. Sys., E","doi":null,"url":null,"internal_url":null},{"position":24,"raw":"[Osherson et al., 1991] Daniel N. Osherson, Joshua Stern, Ormond Wilkie, Michael Stob, and Edward E. Smith. De- fault probability. Cognitive Science, 15(2):251-269, 1991.","title":"Default probability","authors":"N Daniel, Joshua Osherson, Ormond Stern, Michael Wilkie, Edward E Stob, Smith","year":"1991","venue":"Cognitive Science","doi":null,"url":null,"internal_url":null},{"position":25,"raw":"[Palla et al., 2012] Konstantina Palla, David A. Knowles, and Zoubin Ghahramani. An infinite latent attribute model for network data. In Proc. ICML, pages 1607-1614, 2012.","title":"An infinite latent attribute model for network data","authors":"Konstantina Palla, David A Knowles, Zoubin Ghahramani","year":"2012","venue":"Proc. ICML","doi":null,"url":null,"internal_url":null},{"position":26,"raw":"[Roy et al., 2006] Daniel M. Roy, Charles Kemp, Vikash K. Mansinghka, and Joshua B. Tenenbaum. Learning anno- tated hierarchies from relational data. In Proc. NIPS, pages 1185-1192, 2006.","title":"Learning annotated hierarchies from relational data","authors":"M Daniel, Charles Roy, Kemp, K Vikash, Joshua B Mansinghka, Tenenbaum","year":"2006","venue":"Proc. NIPS","doi":null,"url":null,"internal_url":null},{"position":27,"raw":"[Teh et al., 2006a] Yee Whye Teh, Michael I. Jordan, Matthew J. Beal, and David M. Blei. Hierarchical Dirich- let processes. J. Am. Stat. Assoc., 101(476):1566-1581, 2006.","title":"Hierarchical Dirichlet processes","authors":"Teh","year":"2006","venue":"J. Am. Stat. Assoc","doi":null,"url":null,"internal_url":null},{"position":28,"raw":"[Teh et al., 2006b] Yee Whye Teh, David Newman, and Max Welling. A collapsed variational Bayesian inference algo- rithm for latent Dirichlet allocation. In Proc. NIPS, pages 1353-1360, 2006.","title":"A collapsed variational Bayesian inference algorithm for latent Dirichlet allocation","authors":"Teh","year":"2006","venue":"Proc. NIPS","doi":null,"url":null,"internal_url":null},{"position":29,"raw":"[Xu et al., 2006] Zhao Xu, Volker Tresp, Kai Yu, and Hans- Peter Kriegel. Infinite hidden relational models. In Proc. UAI, pages 544-551, 2006.","title":"Infinite hidden relational models","authors":"Zhao Xu, Kai Volker Tresp, Hans-Peter Yu, Kriegel","year":"2006","venue":"Proc. UAI","doi":null,"url":null,"internal_url":null},{"position":30,"raw":"[Zhou, 2015] Mingyuan Zhou. Infinite edge partition models for overlapping community detection and link prediction. In Proc. AISTATS, volume 38, pages 1135-1143, 2015.","title":"Infinite edge partition models for overlapping community detection and link prediction","authors":"Mingyuan Zhou","year":"2015","venue":"Proc. AISTATS","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}