Performance Analysis of Dissimilar Classification Methods using RapidMiner

Abstract

The centre of attention of this research paper is to assess the performance of dissimilar functioning methods of classification technique on data set of Urban Land Cover using RapidMiner software and propose one that gives good performance on mentioned data set of Urban Land Cover. This is an attractive dataset for classifying the high resolution image of urban land cover; may be utilized for several purposes with the tree planning etc. The performance of classification methods C4.5 Decision Tree, Random forest, K-Nearest Neighbor and Naïve Bayes is examined on basis of their accuracy cost of error and kappa values. The class precision and recall is derived from confusion matrix.