Wróblewski J., 2000.Ensembles of classifiers based on approximate reducts. Proc. of CSP 2000 Workshop, Informatik-Bericht Nr. 140, Humboldt-Universitat zu Berlin (2000) vol. 2, pp. 355 - 362. Revised and extended version in: Fundamenta Informaticae 47(3,4), IOS Press(2001), pp. 351 - 360.
ABSTRACT
A problem of improving rough set based expert systems by modifying a notion of reduct is discussed. A notion of approximate reduct is introduced, as well as some proposals of quality measure for such a reduct. A complete classifying system based on approximate reducts is presented and discussed. It is proved that a problem of finding optimal set of classifying agents based on approximate reducts is NP-hard; a genetic algorithm is used to find the suboptimal set. Experimental results show, that the classifying system is effective and relatively fast.