An Optimization Rough Set Boundary Region based Random Forest Classifier

Machine learning is a concerned with the design and development of algorithms. Machine learning is a programming approach to computers to achieve optimization .Classification is the prediction approach in data mining techniques. Decision tree algorithm is the most common classifier to build tree because of it is easier to implement and understand. Attribute selection is a concept by which we want select attributes that are more significant in the given datasets. We proposed a novel hybrid approach combination of Rough Set with Boundary Region and Random Forest algorithm called Rough Set Boundary Region based Random Forest Classifier (RSBRRF Classifier) which is use to deal with uncertainties, vagueness and ambiguity associated with datasets. In this approach, we select significant attributes based on rough set theory with boundary region as an input to random forest classifier for constructing the decision tree is more efficient and scalable approach for classification of various datasets.