Rough Ensemble Classifier: A Comparative Study

摘要: Combining the results of a number individually trained classification systems to obtain more accurate classifier is widely used technique in pattern recognition. In this article, we have introduced rough set based meta (RSM). Theoretical analysis proposed RSM carried out relation Bayes since best classifier. It has been shown that performance at least as good constituent classifier, and if one base classifiers converges then Experimental studies show improves accuracy beats other ensemble approaches by decisive margin, thus demonstrating theoretical results.