A short review of machine learning methods for classifying the outcome of Gestational Diabetes

Diabetes mellitus is a growing problem, especially in developing countries. People suffering from diabetes have an increased risk of developing a number of serious health problems. Consistently high blood glucose levels can lead to serious diseases affecting the heart and blood vessels, eyes, kidney, etc. In addition, people with diabetes also have a higher risk of developing infections. This paper aims to use suitable data mining and classification techniques which include the Logit model, the Probit model, the Classification tree technique, Artificial Neural Networks, Support Vector Machines, Ridge Regression technique and the Least Absolute Shrinkage and Selection Operator(LASSO) in order to determine the best method which can be used to classify the patients as suffering from gestational diabetes or not. The misclassification rate is calculated for different methods and the method having the least misclassification rate is said to be the most suitable to be applied to the given data, which is the PIMA Indians diabetes dataset.