A rule induction algorithm for knowledge discovery and classifcation

Classifcation and rule induction are key topics in the felds of decision making and knowledge discovery. The objective of this study is to present a new algorithm developed for automatic knowledge acquisition in data mining. The proposed algorithm has been named RES-2 (Rule Extraction System). It aims at eliminating the pitfalls and disadvantages of the techniques and algorithms currently in use. The proposed algorithm makes use of the direct rule extraction approach, rather than the decision tree. For this purpose, it uses a set of examples to induce general rules. In this study, 15 datasets consisting of multiclass values with diferent properties and sizes and obtained from the University of California, Irvine, have been used. Classifcation accuracy and rule count have been used to test the proposed method. This method presents an alternative 3-step method to classify categorical, binary, and continuous data by taking advantage of algorithms for data mining classifcation and decision rule generation. The method aims at improving the classifcation accuracy of the algorithms that extract the decision rules. Experimental studies were conducted on the benchmark datasets and the results of the comparisons with some known algorithms for decision rule generation have shown that the proposed method performs classication with a higher accuracy and generates fewer rules.