PERFORMANCE ANALYSIS OF DIFFERENT CLASSIFIERS FOR DIABETES CLASSIFICATION

Conference: Creative Trends in Engineering and Technology
Author(s): Midhila. M, S. Padmavathi Year: 2016
Grenze ID: 02.CTET.2016.1.501_7 Page: 433-437

Abstract

Diabetes during pregnancy is a major issue common among Indian women. Prediction of diabetes based\non the test done during the pregnancy period plays a significant role in the treatment. Results of various clinical test\nconducted during the pregnancy can be considered as parameters for diabetes classification. In this paper, Classifiers\nsuch as Support Vector Machine, probablistic, Tree based and regression are used to predict diabetes based on the\nparameters. The SVM classifiers performed well on the dataset giving the highest accuracy of 78% while regression\nbased classsifier scored a minimum of 65% accuracy.

<< BACK

CTET - 2016