Hybridization of Genetic Algorithm and K Nearest Neighbor Classifier

Machine learning is the study of algorithms that automatically improve their performance with experience. That can provide signi¿cant competitive advantages to many organizations by exploiting the potential of large data volume. Intelligently analyzed data is a valuable resource. At the heart of performance is classi¿cation accuracy in this speci¿ed task. Mostly a crucial problem in machine learning is identifying a representative set of features from which to construct a classi¿cation model for a particular task. The classi¿cation of data is based on the set of data feature used. The feature selection can provide optimizing performance by using Genetic Algorithm and strongly effect in classi¿cation. In making to get improved classi¿cation accuracy, author has taken advantage with using hybrid of machine learning methods rather than use of only machine learning approach. Author proposed Information based distance metric to overwhelm one of the weak points of k nearest neighbor classi¿er. Moreover it provide the comparison of the result of Information based distance metric and Euclidean distance metric on both majority voting and similarity score summing.