Predicting Hypoglycaemia Using Classification

With the advancement of the development in the technology the majority of Type 2 Diabetes Mellitus(T2DM) screening tests in today with the use of multivariate technology and techniques data mining has provided strength and opportunities to the study of dynamic interaction between large variables. Machine learning approach used to predict the early onset of diabetes mellitus (DM). This algorithm has increased the accuracy to forecast the risk of diabetes using classifier models. It predicts the increase of blood glucose whereas deep learning of neural network probabilistic modelling was designed. Since diabetes mellitus is one of the most common chronic condition which has the highest death rate. In order to improve the quality of life of individual with diabetes and to eliminate complication, preventing glycaemic levels from reaching the physiological range in fundamental. The review of 12 paper aim is to provide classification techniques using machine learning methods. It involves the approach provided to collect the pre-processing to obtain relevant characteristics to measure their significant features that is classified through the performance based on precision, sensitivity, specificity and area under the curve and trained through SVM to identify the related features. Out of 30 paper completion 12 paper were nominated to reviewed which mainly focused on prediction model to build into support scheme for diagnosis or integrated with current information system for healthcare.