A Survey of different classification techniques and their comparison using Mc Nemar’s Test

Abstract

Five classification algorithms namely J48, Naive Bayes, K Nearest Neighbour, IBK and Decision Tree are evaluated using Mc Nemar’s test over datasets including both nominal attributes and numeric attributes. It was found that K Nearest Neighbor performed better than the other classification methods for both nominal datasets and numerical datasets. It was also observed that the results of this evaluation confers with two other evaluation metrics used for evaluating classification algorithms or machine learning algorithms, Root Mean Squared Error and Kappa statistic.