PERFORMANCE EVALUATION OF ENSEMBLE LEARNING ALGORITHMS FOR VARIOUS CLASSIFIERS

: In the present scenario with the advancement of digital technology extensive amount of data and information are generated. With such extensive amount of data, powerful approaches are required for data interpretation that help the human being for decision making process. Handling, analyzing, processing and finding relevant information from this extensive amount of data is very interesting and fast growing research area. Data mining is the process of extracting or discovering useful information from the large amount of data and then transforms them in to an understandable form for future use. Data mining offers various methods or techniques that are used to predict the accuracy of various classes of object. This research focus on various ensemble learningh techniques like bagging and boosting also enhance the accuracy of various base classifiers like NaiveBayes, DecisionStump, DecisionTable and J48. All the techniques are compared on the basis of four evaluation parameters like accuracy, precision, recall and root mean squared error. The finding are also supported with justification by conducting an experimental survey at the end of research. RapidMiner tool will be used to perform the simulation using 10 fold cross validation