Prediction of Diabetes and cancer using SVM Classifer and Naïve Bayes

Abstract—Diabetes mellitus is caused due to the increased level of sugar content in the blood. This can cause series complications like kidney failure, stroke, cancer, heart disease and blindness. The early detection and diagnosis, helps to identify and avoid these complications. A number of computerized information systems were designed using different classifiers for predicting and diagnosing diabetes. Selecting proper algorithms for classification clearly increases the accuracy and efficiency of the system. The main objective of this study is to review the benefits of different preprocessing techniques for decision support systems for predicting diabetes which are based on Support Vector Machine (SVM), < Final Year Projects > Naive Bayes classifier and Decision Tree. The preprocessing methods focused on this study are Principal Component Analysis and Discretization. The accuracy variation with and without preprocessing techniques are also evaluated. The tool under consideration is the Weka for this study. The dataset was taken from the University of California, Irvine (UCI) repository of machine learning.