Role of Robustness Measure in Rule Induction
Motoyuki Ohki, Eiji Sekiya, and Masahiro Inuiguchi
Graduate School of Engineering Science, Osaka University
1-3 Machikaneyama, Toyonaka, Osaka 560-8531, Japan
Rough set approaches provide useful tools to induce minimal decision rules from given data. Acquired minimal rules are typically used to build a classifier. However, minimal rules are sometimes used for design knowledge. Specifically, if a new object is designed to satisfy the condition of a minimal rule, it can be classified into a class suggested by the rule. Although we are interested in the goodness of the set of obtained minimal decision rules for the purpose of building a classifier, we are more interested in the goodness of an individual minimal decision rule for design knowledge. In this study, we propose robustness measures as a new type of evaluation index for decision rules. The measure evaluates the extent to which interestingness is preserved after the some conditions are removed. Four numerical experiments are conducted to examine the usefulness of robusetness measures. Decision rules selected by robustness scores are compared with those selected by recall, which is the well-known measure to select good rules. Our results reveal that a different aspect of the goodness of a rule is evaluated by the robustness measure and thus, the robustness measure acts as an independent and complementary index of recall.
-  Z. Pawlak, “Rough Sets,” Int. J. of Computer and Information Sciences, Vol.11, No.5, pp. 341-356, 1982.
-  N. Mori, H. Tanaka, and K. Inoue, “Rough Sets and Kansei,” Kaibundo, 2006 (in Japanese).
-  N. Ytow, D. R. Morse, and D. McL. Roberts, “Rough Set Approximation as Formal Concept,” J. of Advanced Computational Intelligence and Intelligent Informatics (JACIII), Vol.10, No.5, pp. 606-611, 2006.
-  N. Yamaguchi, M. Wu, M. Nakata, and H. Sakai, “Application of Rough Set-Based Information Analysis to Questionnaire Data,” J. of Advanced Computational Intelligence and Intelligent Informatics (JACIII), Vol.18, No.6, pp. 953-961, 2014.
-  M. Inuiguchi and K. Washimi, “Improving Rough Set Rule-Based Classification by Supplementary Rules,” J. of Advanced Computational Intelligence and Intelligent Informatics (JACIII), Vol.19, No.6, pp. 747-758, 2015.
-  L. Geng and H. J. Hamilton, “Interestingness Measure for Data Mining: A Survey,” ACM Computing Surveys, Vol.38, No.9, pp. 1-32, 2006.
-  K. Macgarry, “A Survey of Interestingness Measure for Knowledge Discovery,” The Knowledge Engineering Review, pp. 1-24, 2005.
-  S. Greco, R. Slowinski, and I. Szczch, “Properties of Rule Interestingness Measure and Alternative Approaches to Normalization of Measures,” Information Sciences, Vol.216, pp. 1-16, 2012.
-  W. Ziarko, “Variable Precision Rough Set Model,” J. of Computer and System Sciences, Vol.46, No.1, pp. 39-59, 1993.
-  P. Lenca, B. Vaillant, and S. Lallich, “On the Robustness of Association Rules,” Proc. of 2006 IEEE Conf. on Cybernetics and Intelligent Systems, pp. 596-601, 2006.
-  Y. L. Bras, P. Meyer, P. Lenca, and S. Lallich, “A Robustness Measure of Association Rules,” Machine Learning and Knowledge Discovery in Databases, LNCS 6322, pp. 227-242, 2010.
-  M. Ohki and M. Inuiguchi, “Robustness Measure of Decision Rules,” Proc. of RSKT, LNAI 8171, pp. 166-177, 2013.
-  J. W. Grzymala-Busse, “MLEM2: Discretization During Rule Induction,” Proc. of the IIPWM 2003, pp. 499-508, 2003.
-  UCI Machine Learning Repository, http://archive.ics.uci.edu/ml/ [Accessed February 1, 2016]
-  J. W. Grzymala-Busse, “LERS: A System for Learning from Examples Based on Rough Sets,” Intelligent Decision Support: Handbook of Applications and Advance of the Rough Sets Theory, Kluwer Academic Publishers, pp. 3-18, 1992.