The paper addresses the problem of computing short reducts in information/decision tables. Reducts in general, and short reducts in particular, may be usefully applied for consitency preserving data size reduction but are hard to find because of high theoretical complexity of the problem. Practical experiments demonstrate, however, that reducts may be successfully computed for many real life data sets using some advanced algorithms. This paper reports on a series of experiments designed to verify not the theoretical complexity but the practical behaviour of algorithms for reduct computation in the average case. In particular, the problem of computing short reducts is sotved by presenting a new algorithm, which is based on the notion of discernibility matrix. All the results of the experiments reported in this paper have been obtained for real-life data sets:
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.