Feature Selection Based on Confirmation-Theoretic Rough Sets
As an important part of data preprocessing in machine learning and data mining, feature selection, also known as attribute reduction in rough set theory, is the process of choosing the most informative subset of features. Rough set theory has been used as such a tool with much success. The main objective of this paper is to propose a feature selection procedure based on a special group of probabilistic rough set models, called confirmation-theoretic rough set model(CTRS). Different from the existing attribute reduction methods, the definition of positive features is based on Bayesian confirmation measures. The proposed method is further divided into two categories based on the qualitative and quantitative nature of the underlying rough set models. This study provides new insights into the problem of attribute reduction.
Keywordsfeature selection attribute reduction probabilistic rough set confirmation-theoretic rough set
Unable to display preview. Download preview PDF.
- 1.Bayes, T., Price, R.: An essay towards solving a problem in the doctrine of chance. By the late Rev. Mr. Bayes, communicated by Mr. Price, in a letter to John Canton, M. A. and F. R. S. Philosophical Transactions of the Royal Society of London 53, 370–418 (1763)Google Scholar
- 5.Pawlak, Z.: Rough Sets. Theoretical Aspects of Reasoning about Data. Kluwer Academic Publishers, Dordrecht (1991)Google Scholar
- 6.Pawlak, Z., Skowron, A.: Rough membership functions. In: Yager, R.R., Fedrizzi, M., Kacprzyk, J. (eds.) Advances in the Dempster-Shafer Theory of Evidence, pp. 251–271. John Wiley and Sons, New York (1994)Google Scholar
- 9.Ślęzak, D., Ziarko, W.: Bayesian rough set model. In: Proceedings of FDM 2002, pp. 131–135 (2002)Google Scholar
- 12.Yao, Y.Y., Wong, S.K.M., Lingras, P.: A decision-theoretic rough set model. In: Ras, Z.W., Zemankova, M., Emrich, M.L. (eds.) Methodologies for Intelligent Systems 5, pp. 17–24. North-Holland, New York (1990)Google Scholar