A rough set based hybrid approach for classification

Uncertainty defined as a situation with inadequate information can be of three types: inexactness, unreliability, and ignorance, and not merely the absence of knowledge. However, uncertainty can prevail in cases where a considerable amount of information is available. In this regard, Rough Set Theory (RST) is a new mathematical model that deals with uncertain information. Thus, in this paper, we propose a new hybrid approach that combines RST together with Machine Learning (ML) algorithm to improve the classification performance efficiently. We use three ML algorithms, e.g., K-Nearest Neighbors (KNN), Naive Bayes (NB), and Support Vector Machine (SVM), for experimental validation. The results confirm that the proposed method can achieve a remarkable classification performance.