Effective feature selection for supervised learning using genetic algorithm

Feature selection is an effective technique for dimensionality reduction and an essential step in successful data mining applications. It is a process of selecting a subset of features from the candidate set of features according to certain criteria. The main goal of supervised learning is finding feature subset that produces higher classification accuracy. The proposed method is to select an optimal set of features by using Genetic Algorithm that has been done in parallel by using Mapreduce framework. The resultant features will be given it to the K-Nearest Neighbour classifier. The fitness of accuracy will be evaluated using K-NN. Results are validated using the Datasets taken from the UCI machine learning repository. The results indicate that the Parallel GA produces high accuracy than other methods.