Abstract
Approaches to support an interpretation of rules induced from imbalanced data are discussed. In this paper, the rule learning algorithm BRACID dedicated to class imbalance is considered. As it may induce too many rules, which hinders their interpretation, their filtering is applied. We introduce three different strategies, which aim at selecting rules having good descriptive characteristics. The strategies are based on combining Bayesian confirmation measures with rule support, which have not yet been studied in the class imbalance context. Experimental results show that these strategies reduce the number of rules and improve values of rule interestingness measures at the same time, without considerable losses of prediction abilities, especially for the minority class.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsNotes
- 1.
\(c_3(H,E)=A(H,E)Z(H,E)\) in case of confirmation and
\(c_3(H,E)=-A(H,E)Z(H,E)\) in case of disconfirmation
where
\(Z(H,E)=1-P( \lnot H|E) \div P(\lnot H)\) in case of confirmation and
\(Z(H,E)=P(H|E) \div P(H)-1\) in case of disconfirmation;
\(A(H,E)=[P(E|H)-P(E)]\div [1-P(E)]\) in case of confirmation and
\(A(H,E)=[P(H)-P(H| \lnot E)] \div [1-P(H)]\) in case of disconfirmation.
- 2.
For simplicity we will further use a notation of a rule as R instead of (H, E) in symbols of measures.
- 3.
More detailed experimental results, including also the coverage option are provided at the page http://www.cs.put.poznan.pl/iszczech/publications/nfmcp-2016.html.
References
Bayardo, R., Agrawal, R.: Mining the most interesting rules. In: Proceedings of 5th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, pp. 145–154 (1999)
Christensen, D.: Measuring confirmation. J. Philos. 96, 437–461 (1999)
Fitelson, B.: The plurality of Bayesian measures of confirmation and the problem of measure sensitivity. Philos. Sci. 66, 362–378 (1999)
Freitas, A.: On rule interestingness measures. Knowl.-Based Syst. 12, 309–315 (1999)
Furnkranz, J., Gamberger, D., Lavrac, N.: Foundations of Rule Learning. Springer, Berlin (2012). doi:10.1007/978-3-540-75197-7
Gamberger, D., Lavrac, N.: Expert-guided subgroup discovery: methodology and application. J. Artif. Int. Res. 17(1), 501–527 (2002)
Geng, L., Hamilton, H.: Interestingness measures for data mining: a survey. ACM Comput. Surv. 38(3), 9 (2006)
Glass, D.: Confirmation measures of association rule interestingness. Knowl.-Based Syst. 44, 65–77 (2013)
Greco, S., Slowinski, R., Szczech, I.: Properties of rule interestingness measures and alternative approaches to normalization of measures. Inf. Sci. 216, 1–16 (2012)
Greco, S., Slowinski, R., Szczech, I.: Measures of rule interestingness in various perspectives of confirmation. Inf. Sci. 346, 216–235 (2016)
He, H., Yungian, M. (eds.): Imbalanced Learning. Foundations, Algorithms and Applications. IEEE - Wiley, Hoboken (2013)
Heravi, M., Zaiane, O.R.: A study on interestingness measures for associative classifiers. In: Proceedings of ACM-SAC 2010 Conference Track on Data Mining, pp. 1040–1047 (2010)
Lavrač, N., Flach, P., Zupan, B.: Rule evaluation measures: a unifying view. In: Džeroski, S., Flach, P. (eds.) ILP 1999. LNCS (LNAI), vol. 1634, pp. 174–185. Springer, Heidelberg (1999). doi:10.1007/3-540-48751-4_17
Lenca, P., Vaillant, B., Meyer, P., Lallich, S.: Associations rule interestingness measures: experimental and theoretical studies. In: Guillet, F., Hamilton, H.J. (eds.) Quality Measures in Data Mining. SCI, vol. 43, pp. 51–76. Springer, Heidelberg (2007). doi:10.1007/978-3-540-44918-8_3
McGarry, K.: A survey of interestingness measures for knowledge discovery. Knowl. Eng. Rev. 20(1), 39–61 (2005)
Napierala, K., Stefanowski, J.: BRACID: a comprehensive approach to learning rules from imbalanced data. J. Intell. Inf. Syst. 39(2), 335–373 (2012)
Napierala, K., Stefanowski, J.: Types of minority class examples and their influence on learning classifiers from imbalanced data. J. Intell. Inf. Syst. 46(3), 563–597 (2016)
Napierala, K., Stefanowski, J.: Post-processing of BRACID rules induced from imbalanced data. Fundam. Inform. 148(1–2), 51–64 (2016)
Nozick, R.: Philosophical Explanations. Clarendon Press, Oxford (1981)
Stefanowski, J., Vanderpooten, D.: Induction of decision rules in classification and discovery-oriented perspectives. Int. J. Intell. Syst. 16(1), 13–28 (2001)
Acknowledgement
The research was supported by NCN grant DEC-2013/11/B/ST6/00963.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2017 Springer International Publishing AG
About this paper
Cite this paper
Napierała, K., Stefanowski, J., Szczȩch, I. (2017). Increasing the Interpretability of Rules Induced from Imbalanced Data by Using Bayesian Confirmation Measures. In: Appice, A., Ceci, M., Loglisci, C., Masciari, E., Raś, Z. (eds) New Frontiers in Mining Complex Patterns. NFMCP 2016. Lecture Notes in Computer Science(), vol 10312. Springer, Cham. https://doi.org/10.1007/978-3-319-61461-8_6
Download citation
DOI: https://doi.org/10.1007/978-3-319-61461-8_6
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-61460-1
Online ISBN: 978-3-319-61461-8
eBook Packages: Computer ScienceComputer Science (R0)