Learning efficient rulesets from fuzzy data with a genetic algorithm
The aim of this paper is to demonstrate the feasibility of fuzzy measures of subsethood in learning from examples. Using the relationship between (fuzzy) set containment and (fuzzy) logical implication, a method of generating if-then rules that describe a fuzzy dataset is given. In order to obtain an efficient subset of the generated rules, we apply a simple genetic algorithm.
The proposed method is illustrated with a fuzzified well-known learning set. The results on this set clearly improve other approaches.
Unable to display preview. Download preview PDF.
- 2.Botana, F.: Deriving fuzzy subsethood measures from violations of the implication between elements. Lec. Notes Artif. Intel. 1415 (1998) 234–243Google Scholar
- 3.Botana, F.: Learning rules from fuzzy datasets. Proc. 5th Europ. Congress on Intel. Techniques and Soft Comp., Aachen, Germany (1997) 1109–1113Google Scholar
- 10.Michalski, R.S., Chilausky, R.L.: Learning by being told and learning from examples. Int. J. Policy Anal. Inform. Systems 4 2 (1980) 125–160Google Scholar
- 11.Michalski, R.S., Mozetic, I., Hong, J., Lavrac, N.: Multi-purpose incremental learning system AQ15 and its testing application to three medical domains. Proc. AAAI-86, 1041–1045. Morgan Kaufmann, Philadelphia (1986)Google Scholar
- 12.Michalski, R.S., Kaufman, K.: Learning from inconsistent or noisy data: the ENIL/AQ18 approach. Lec. Notes Artif. Intel. (1999), to appearGoogle Scholar
- 14.Quinlan, J.R.: C4.5: Programs for Machine Learning. Morgan Kaufmann, San Mateo (1993)Google Scholar
- 15.Quinlan, J.R.: Induction of decision trees. Mach. Learning 1(1) (1986) 81–106Google Scholar
- 19.Weber, R.: Fuzzy-ID3: a class of methods for automatic knowledge acquisition. Proc. 2nd Int. Conf. on Fuzzy Logic,Iizuka, Japan, (1992) 265–268Google Scholar