Comparison of affinity degree classification with four different classifiers in several data sets

Abstract

The affinity notion has been widely used in research fields. Thus, in this research, affinity is employed to find the degree between two data sets and classify through prediction. But, as Affinity Degree (AD) classification is a new technique, the comparison with different classification types is needed to test the compatibility technique. Herein, this study compares various machine learning techniques and determines the most efficient classification technique based on the data set. Four different classification algorithms, K-Nearest Neighbour (KNN), Naive Bayes (NB), Decision Tree (J48), and Support Vector Machine (SVM), were used as other techniques to compare with AD classification. Three different data sets, breast cancer, acute inflammation, and iris plant, were used for experiment purposes. The results show J48 has the best rate in performance measures compare to the other four classifiers. However, the results of AD classification show the significance that more studies can improve it.