Predict Blood Glucose Level Using Least-Squares Support Vector Machine Regression and Fuzzy Clustering

Patients with diabetes should continuously monitor their blood glucose levels and adjust insulin levels according to the normal range. Therefore, they should use methods to reduce the number of blood glucose measurements and prevent the subsequent effects of these measurements in order to predict possible dangerous complications that can be considered as the most important step to diagnose, control and even treat these complications. This paper presents a new algorithm to predict blood glucose levels using a least-squares support vector machine regression based on Fuzzy C-Means Clustering. The proper functioning of this method has been shown to predict blood glucose levels over other methods by testing and validating the proposed algorithm with the real data obtained from India s Diabetes Database in UCI Machine Learning Laboratory

Keywords:

Regression , least squares support vector machines , fuzzy clustering