Classification of Diabetes Mellitus (DM) using Machine Learning Algorithms

Diabetes Mellitus (DM) is one of the most prevalent disease in the world today which
is associated by having high glucose level in body either due to inadequate production
of insulin or the body cell’s not responding towards the produced insulin. Data mining
and machine learning techniques can be extremely useful in classification of DM
considering the need to have a shift from current traditional method which uses sharp
needles to draw blood towards a non – invasive method. The objective of this study is
to perform DM classification using various machine learning algorithms using Weka
as a tool. In this paper, single classifiers such as Support Vector Machine, Naïve Bayes,
Bayes Net, Decision Stump, k – Nearest Neighbors, Logistic Regression, Multilayer
Perceptron and Decision Tree is experimented. Apart from that, ensemble methods
such as bagging, adaptive boosting using AdaBoostM1, hybrid classifier using
combinations of Random Forest with various base classifiers and ensemble algorithm
which is the Random Forest has also been studied. In this research, it was found that
performance of ensemble method using hybrid classifier of Random Forest – Bayes
Net model was found as the best DM classification model with an accuracy of 83.91%
using the Pima Indian Diabetes Dataset (PIDD) out beating all the other classification
algorithms. Whereas for the German Frankfurt dataset, best DM classification model
was found using Random Forest algorithm with an accuracy of 98.77%.