A Comparative Analysis of Various Machine Learning Methods to Predict Diabetes Mellitus
Authors
JAMALUDDIN MIR, AYESHA ASLAM, KALSOOM AYYAZ, GOHAR ZAMAN, WAJID KHAN and MUHAMMAD AYYAZ
Abstract
The recent advancements in the field of health sciences have produced substantial amount of data such as clinical information that is generated by patient records which is used in AI applications for better diagnosis and predictions. Diabetes belongs to a group of metabolic disorders that affects 422 million people worldwide. This is primarily due to lack of predictive and forecasting measures. Research on several aspects of diabetes has generated huge amounts of data which makes it suitable for application of AI based methods. Presently, several methods have been used for predicting diabetes on the basis of certain factors. However, results of this study show that Support Vector Machine (SVM) and Linear regression when combined with statistical methods, provide much better results compared to AI methods.
Keywords: Diabetes, Machine Learning, Support Vector Machine, PIMA Dataset, Prediction.
