Bipartite Graph Representation of Multiple Decision Table Classifiers
In this paper, we consider the two-class classification problem, a significant issue in machine learning. From a given set of positive and negative samples, the problem asks to construct a classifier that predicts the classes of future samples with high accuracy. For this problem, we have studied a new visual classifier in our previous works, which is constructed as follows: We first create several decision tables and extract a bipartite graph structure (called an SE-graph) between the given set of samples and the set of created decision tables. We then draw the bipartite graph as a two-layered drawing by using an edge crossing minimization technique, and the resulting drawing acts as a visual classifier. We first describe our background and philosophy on such a visual classifier, and then consider improving its classification accuracy. We demonstrate the effectiveness of our methodology by computational studies using benchmark data sets, where the new classifier outperforms our older version, and is competitive even with such standard classifiers as C4.5 or LibSVM.
Unable to display preview. Download preview PDF.
- 4.Haraguchi, K., Hong, S., Nagamochi, H.: Visual analysis of hierarchical data using 2.5D drawing with minimum occlusion. In: IEEE PacificVis 2008 (2008) (poster session)Google Scholar
- 5.Haraguchi, K., Hong, S., Nagamochi, H.: Visual analysis of hierarchical data using 2.5D drawing with minimum occlusion. Technical Report 2009-010, Department of Applied Mathematics and Physics, Graduate School of Informatics, Kyoto University, Japan (2009)Google Scholar
- 7.Dwork, C., Kumar, R., Naor, M., Sivakumar, D.: Rank aggregation methods for the web. In: WWW10, pp. 613–622. ACM, New York (2001)Google Scholar
- 9.Haraguchi, K., Hong, S., Nagamochi, H.: Classification by ordering data samples. RIMS Kokyuroku 1644, 20–34 (2009)Google Scholar
- 10.Haraguchi, K., Hong, S., Nagamochi, H.: Classification via visualization of sample-feature bipartite graphs. Technical Report 2009-011, Department of Applied Mathematics and Physics, Graduate School of Informatics, Kyoto University, Japan (2009)Google Scholar
- 12.Haraguchi, K., Hong, S., Nagamochi, H.: Visualization can improve multiple decision table classifiers. In: Proc. of MDAI (to appear, 2009)Google Scholar
- 13.Quinlan, J.R.: C4.5: Programs for Machine Learning. Morgan Kaufmann, San Francisco (1993)Google Scholar
- 14.Chang, C.C., Lin, C.J.: LIBSVM: a library for support vector machines (2001), http://www.csie.ntu.edu.tw/~cjlin/libsvm
- 24.Asuncion, A., Newman, D.: UCI Machine Learning Repository, University of California, Irvine, School of Information and Computer Sciences (2007), http://www.ics.uci.edu/~mlearn/MLRepository.html
- 26.Weiss, S.M., Kulikowski, C.A.: Computer Systems that Learn: Classification and Prediction Methods from Statistics, Neural Nets, Machine Learning, and Expert Systems. Morgan Kaufmann, San Francisco (1991)Google Scholar