Rule Extraction Framework Using Rough Sets and Neural Networks

This paper deals with the simplification of classification rules for data mining using rough sets theory combined with neural networks. In the attribute reduction process, the proposed approach generates minimal reduct and minimum number of rules with high accuracy. Experimental results with sample data sets in UCI repository show that this method gives a good performance in getting concise and accurate rules.

[1]  Byeong Seok Ahn,et al.  The integrated methodology of rough set theory and artificial neural network for business failure prediction , 2000 .

[2]  Ramin Yasdi,et al.  Combining Rough Sets Learning- and Neural Learning-method to deal with uncertain and imprecise information , 1995, Neurocomputing.

[3]  Andrew Kusiak,et al.  Rough set theory: a data mining tool for semiconductor manufacturing , 2001 .

[4]  Jerzy W. Grzymala-Busse,et al.  Rough Sets , 1995, Commun. ACM.

[5]  Li Xiangyang,et al.  A novel self-optimizing approach for knowledge acquisition , 2000 .

[6]  Nick Cercone,et al.  Generation of Multiple Knowledge from Databases Based on Rough Sets Theory , 1997 .

[7]  Yindong Ji,et al.  Entropy based uncertainty measures for classification rules with inconsistency tolerance , 2000, Smc 2000 conference proceedings. 2000 ieee international conference on systems, man and cybernetics. 'cybernetics evolving to systems, humans, organizations, and their complex interactions' (cat. no.0.

[8]  Sankar K. Pal,et al.  Staging of cervical cancer with soft computing , 2000, IEEE Transactions on Biomedical Engineering.

[9]  Janusz A. Starzyk,et al.  A Mathematical Foundation for Improved Reduct Generation in Information Systems , 2000, Knowledge and Information Systems.

[10]  Usama M. Fayyad,et al.  Multi-Interval Discretization of Continuous-Valued Attributes for Classification Learning , 1993, IJCAI.

[11]  Sankar K. Pal,et al.  Rough-Fuzzy MLP: Modular Evolution, Rule Generation, and Evaluation , 2003, IEEE Trans. Knowl. Data Eng..