ANALYSIS OF CLASSIFICATION ALGORITHMS ON DIFFERENT DATASETS

Purpose. Data mining is the forthcoming research area to solve different problems and classification is one of main problem in the field of data mining. In this paper, we use two classification algorithms J48 and Sequential Minimal Optimization alias SMO of the Weka interface. Methodology. It can be used for testing several datasets. The performance of J48 and Sequential Minimal Optimization has been analyzed to choose the better algorithm based on the conditions of the datasets. The datasets have been chosen from UCI Machine Learning Repository. Findings. Algorithm J48 is based on C4.5 decision-based learning and algorithm Sequential Minimal Optimization uses the Support Vector Machine approach for classification of datasets. When comparing the performance of both algorithms we found Sequential Minimal Optimization is better algorithm in most of the cases. Originality. This is the first implemented research work up to my knowledge, data sets classification problem handled in data mining using SMO with Weka interface