Collative Study of Classifiers in Patteren Recognition
Pattern recognition which is a field of machine learning has evolved from artificial intelligence. In turn it helps users to make decisions and recognize various patterns. The two fields of pattern recognition are classification and regression. Classification a form of supervised learning helps them to classify the training data into correctly labeled dataset. Classification trees also known as classifiers are used to classify data into expected class. Random forest and REPtree are two such classifiers. Random forest is ensemble of predictor variables which classifies the data in input vector into correctly identified classes. REPTree is fast learning regression tree which is suitable for classifying numerical values. This paper presents algorithm and flow chart for classification using random forest and REPTree. Each of these classifiers have their own advantages and disadvantages experiments are conducted using WEKA3.7 open source tool and IRIS database set.