{"type":"ScholarlyArticle","title":"Handling Noise in Boolean Matrix Factorization","authors":[{"name":"Radim Belohlavek","slug":"radim-belohlavek","given":"Radim","family":"Belohlavek","affiliation":"Dept. of Computer Science , Palacký University Olomouc , Czech Repiblic"},{"name":"Martin Trnecka","slug":"martin-trnecka","given":"Martin","family":"Trnecka","affiliation":"Dept. of Computer Science , Palacký University Olomouc , Czech Repiblic"}],"abstract":"We critically examine and point out weaknesses of the existing considerations in Boolean matrix factorization (BMF) regarding noise and the algorithms' ability to deal with noise. We argue that the current understanding is underdeveloped and that the current approaches are missing an important aspect. We provide a new, quantitative way to assess the ability of an algorithm to handle noise. Our approach is based on a common-sense definition of robustness requiring that the computed factorizations should not be affected much by varying the noise in data. We present an experimental evaluation of several existing algorithms and compare the results to the observations available in the literature. In addition to providing justification of some properties claimed in the literature without proper justification, our experiments reveal properties which were not reported as well as properties which counter certain claims made in the literature. Importantly, our approach reveals a line separating robust-to-noise from sensitive-to-noise algorithms, which has not been revealed by the previous approaches.","doi":"10.24963/ijcai.2017/198","url":"https://ijcai.org/proceedings/2017/198","pdf":"https://ijcai.org/proceedings/2017/0198.pdf","markdown":"https://ijcai.org/proceedings/2017/198.md","bibtex":"https://ijcai.org/proceedings/2017/bibtex/198","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":198,"pages":{"first":1433,"last":1439},"keywords":["Machine Learning: Data Mining","Machine Learning: Feature Selection/Construction"],"videos":[],"references":[{"position":1,"raw":"[Belohlavek and Trnecka, 2015] R. Belohlavek and M. Tr- necka. From-below approximations in Boolean matrix fac- torization: Geometry and new algorithm. Journal of Com- puter and System Sciences, 81(8):1678-1697, June 2015.","title":"From-below approximations in Boolean matrix factorization: Geometry and new algorithm","authors":"R Belohlavek, M Trnecka","year":"2015","venue":"Journal of Computer and System Sciences","doi":null,"url":null,"internal_url":null},{"position":2,"raw":"[Belohlavek and Vychodil, 2009] R. Belohlavek and V. Vy- chodil. Discovery of optimal factors in binary data via a novel method of matrix decomposition. Journal of Com- puter and System Sciences, 76(1):3-20, May 2009.","title":"Discovery of optimal factors in binary data via a novel method of matrix decomposition","authors":"R Belohlavek, V Vychodil","year":"2009","venue":"Journal of Computer and System Sciences","doi":null,"url":null,"internal_url":null},{"position":3,"raw":"[Dixon, 1992] W. J. Dixon. BMDP statistical software man- ual: To accompany BMDP release 7. University of Cali- fornia Press, 1992.","title":"BMDP statistical software manual: To accompany BMDP release 7","authors":"W J Dixon","year":"1992","venue":null,"doi":null,"url":null,"internal_url":null},{"position":4,"raw":"[Ene et al., 2008] A. Ene, W. Horne, N. Milosavljevic, P. Rao, R. Schreiber, and R. E. Tarjan. Fast exact and heuristic methods for role minimization problems. In Pro- ceedings of the 13th ACM Symposium on Access Control Models and Technologies, pages 1-10, Estes Park, Col- orado, USA, June 2008. ACM.","title":"Fast exact and heuristic methods for role minimization problems","authors":"A Ene, W Horne, N Milosavljevic, P Rao, R Schreiber, R E Tarjan","year":"2008","venue":"Proceedings of the 13th ACM Symposium on Access Control Models and Technologies","doi":null,"url":null,"internal_url":null},{"position":5,"raw":"[Geerts et al., 2004] F. Geerts, B. Goethals, and T. Mielikainen. Tiling databases. In Proceedings of the 7th International Conference on Discovery Science, pages 278-289, Berlin, Heidelberg, Oct 2004. Springer Berlin Heidelberg.","title":"Tiling databases","authors":"F Geerts, B Goethals, T Mielikainen","year":"2004","venue":"Proceedings of the 7th International Conference on Discovery Science","doi":null,"url":null,"internal_url":null},{"position":6,"raw":"[Gupta et al., 2008] R. Gupta, G. Fang, B. Field, M. Stein- bach, and V. Kumar. Quantitative evaluation of approxi- mate frequent pattern mining algorithms. In Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '08, pages 301-309, New York, NY, USA, 2008. ACM.","title":"Quantitative evaluation of approximate frequent pattern mining algorithms","authors":"R Gupta, G Fang, B Field, M Steinbach, V Kumar","year":"2008","venue":"Proceedings of the 14th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, KDD '08","doi":null,"url":null,"internal_url":null},{"position":7,"raw":"[Karaev et al., 2015] S. Karaev, P. Miettinen, and J. Vreeken. Getting to know the unknown unknowns: Destructive- noise resistant Boolean matrix factorization. In Proceed- ings of the 2015 SIAM International Conference on Data Mining, SDM 2015, pages 325-333, 2015.","title":"Getting to know the unknown unknowns: Destructivenoise resistant Boolean matrix factorization","authors":"S Karaev, P Miettinen, J Vreeken","year":"2015","venue":"Proceedings of the 2015 SIAM International Conference on Data Mining","doi":null,"url":null,"internal_url":null},{"position":8,"raw":"[Kim, 1982] K. H. Kim. Boolean matrix theory and appli- cations. Monographs and textbooks in pure and applied mathematics. Dekker, 1982.","title":"Boolean matrix theory and applications","authors":"K H Kim","year":"1982","venue":"Monographs and textbooks in pure and applied mathematics","doi":null,"url":null,"internal_url":null},{"position":9,"raw":"[Lu et al., 2008] H. Lu, J. Vaidya, and V. Atluri. Optimal Boolean matrix decomposition: Application to role engi- neering. In Proceedings of the 2008 IEEE 24th Interna- tional Conference on Data Engineering, ICDE '08, pages 297-306, Washington, DC, USA, 2008. IEEE Computer Society.","title":"Optimal Boolean matrix decomposition: Application to role engineering","authors":"H Lu, J Vaidya, V Atluri","year":"2008","venue":"Proceedings of the 2008 IEEE 24th International Conference on Data Engineering, ICDE '08","doi":null,"url":null,"internal_url":null},{"position":10,"raw":"[Lucchese et al., 2010] C. Lucchese, S. Orlando, and R. Perego. Mining top-k patterns from binary datasets in presence of noise. In SIAM DM 2010, pages 165-176.","title":"Mining top-k patterns from binary datasets in presence of noise","authors":"C Lucchese, S Orlando, R Perego","year":"2010","venue":"SIAM DM","doi":null,"url":null,"internal_url":null},{"position":11,"raw":"[Lucchese et al., 2014] C. Lucchese, S. Orlando, and R. Perego. A unifying framework for mining approximate top-k binary patterns. IEEE Transactions on Knowledge and Data Engineering, 26(12):2900-2913, Dec 2014.","title":"A unifying framework for mining approximate top-k binary patterns","authors":"C Lucchese, S Orlando, R Perego","year":"2014","venue":"IEEE Transactions on Knowledge and Data Engineering","doi":null,"url":null,"internal_url":null},{"position":12,"raw":"[Miettinen and Vreeken, 2011] P. Miettinen and J. Vreeken. Model order selection for Boolean matrix factorization. In Proceedings of the 17th ACM SIGKDD International Con- ference on Knowledge discovery and data mining, pages 51-59. ACM, 2011.","title":"Model order selection for Boolean matrix factorization","authors":"P Miettinen, J Vreeken","year":"2011","venue":"Proceedings of the 17th ACM SIGKDD International Conference on Knowledge discovery and data mining","doi":null,"url":null,"internal_url":null},{"position":13,"raw":"[Miettinen and Vreeken, 2014] P. Miettinen and J. Vreeken. Mdl4bmf: Minimum description length for Boolean ma- trix factorization. ACM Transactions on Knowledge Dis- covery from Data (TKDD), 8(4):18, 2014.","title":"Mdl4bmf: Minimum description length for Boolean matrix factorization","authors":"P Miettinen, J Vreeken","year":"2014","venue":"ACM Transactions on Knowledge Discovery from Data (TKDD)","doi":null,"url":null,"internal_url":null},{"position":14,"raw":"[Miettinen et al., 2008] P. Miettinen, T. Mielikainen, A. Gio- nis, G. Das, and H. Mannila. The discrete basis problem. IEEE Transactions on Knowledge and Data Engineering, 20(10):1348-1362, Oct 2008.","title":"The discrete basis problem","authors":"P Miettinen, T Mielikainen, A Gionis, G Das, H Mannila","year":"2008","venue":"IEEE Transactions on Knowledge and Data Engineering","doi":null,"url":null,"internal_url":null},{"position":15,"raw":"[Miettinen, 2009] P. Miettinen. Matrix decomposition meth- ods for data mining: Computational complexity and algo- rithms. PhD thesis, Helsingin yliopisto, 2009.","title":"Matrix decomposition methods for data mining: Computational complexity and algorithms","authors":"P Miettinen","year":"2009","venue":null,"doi":null,"url":null,"internal_url":null},{"position":16,"raw":"[Nau et al., 1978] D. S. Nau, G. Markowsky, M. A. Wood- bury, and A. D. Bernard. A mathematical analysis of human leukocyte antigen serology. Mathematical Bio- sciences, 40(3-4):243-270, 1978.","title":"A mathematical analysis of human leukocyte antigen serology","authors":"D S Nau, G Markowsky, M A Woodbury, A D Bernard","year":"1978","venue":"Mathematical Biosciences","doi":null,"url":null,"internal_url":null},{"position":17,"raw":"[Nau, 1976] D. S. Nau. Specificity covering. Technical re- port, Tech. Rep. CS-1976-7, Duke University, 1976.","title":"Specificity covering","authors":"D S Nau","year":"1976","venue":null,"doi":null,"url":null,"internal_url":null},{"position":18,"raw":"[Stockmeyer, 1975] L. J. Stockmeyer. The set basis problem is NP-complete. Research reports. IBM Thomas J. Watson Research Division, 1975.","title":"The set basis problem is NP-complete","authors":"L J Stockmeyer","year":"1975","venue":null,"doi":null,"url":null,"internal_url":null},{"position":19,"raw":"[Tatti and Vreeken, 2012] N. Tatti and J. Vreeken. Compar- ing apples and oranges: Measuring differences between exploratory data mining results. Data Mining and Knowl- edge Discovery, 25(2):173-207, 2012.","title":"Comparing apples and oranges: Measuring differences between exploratory data mining results","authors":"N Tatti, J Vreeken","year":"2012","venue":"Data Mining and Knowledge Discovery","doi":null,"url":null,"internal_url":null},{"position":20,"raw":"[Vaidya et al., 2007] J. Vaidya, V. Atluri, and Q. Guo. The role mining problem: Finding a minimal descriptive set of roles. In Proceedings of the 12th ACM Symposium on Access Control Models and Technologies, pages 175-184. ACM, 2007.","title":"The role mining problem: Finding a minimal descriptive set of roles","authors":"J Vaidya, V Atluri, Q Guo","year":"2007","venue":"Proceedings of the 12th ACM Symposium on Access Control Models and Technologies","doi":null,"url":null,"internal_url":null},{"position":21,"raw":"[Xiang et al., 2011] Y. Xiang, R. Jin, D. Fuhry, and F. F. Dra- gan. Summarizing transactional databases with overlapped hyperrectangles. Data Mining and Knowledge Discovery, 23(2):215-251, Sep 2011.","title":"Summarizing transactional databases with overlapped hyperrectangles","authors":"Y Xiang, R Jin, D Fuhry, F F Dragan","year":"2011","venue":"Data Mining and Knowledge Discovery","doi":null,"url":null,"internal_url":null}],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}