We would like to announce a new symbolic, inductive learning algorithm (ILA-2) that can be used in data mining. ILA-2 is a rule induction algorithm that generates classifiers in the form of set of rules when provided with examples of a concept. Original version of ILA-2, namely ILA, was developed for noise free domains. Later, we have enhanced ILA by a new evaluation metric that handles uncertainty in the data. The new version of the algorithm is implemented on top of the C4.5 source code. Executable code of the system for Windows platforms is available. Both the source code and executable code for other platforms are also available. Our URL for further information and download is: http://cmpent.eng.emu.edu.tr/ila2/ Dr. Mehmet R. Tolun, Professor Dept. of Computer Engineering Eastern Mediterranean University Gazimagusa T.R.N.C. (via Mersin 10, Turkey) tolun@ieee.org
Copyright © 1999 KDnuggets