COMPARATIVE STUDY OF VARIOUS CLASSIFIERS ON HYPOTHYROID DATA USING WEKA

Abstract

In this paper, WEKA tool has been used to evaluate the performance of various classifiers on a dataset
namely, hypothyroid, to come out with the optimum classifier, for a particular application. Hypothyroid is an
imbalance dataset which contains 28 attributes and 3772 instances. A Classifier plays an important role in any
machine learning application. There are various performance analysis measures that can be used to evaluate the
efficiency of a classifier. In this paper, Naive Bayes, J48, IBK, Vote, Logistic and Random Forest classifiers along
with their combination have been implemented and analysed using WEKA. Accuracy of individual classifiers along
with the accuracy obtained while using the combination of these classifiers have been measured and evaluated. Use of
these hybrid approach helped in combining different classifiers to get the best results.