Early Prediction of Diabetes Diseases Using Optimized Weed Technique
Diabetes is a dreadful disease in which a person suffers from excessive blood glucose in the body because the development of insulin is inadequate or because the body cells does not respond appropriately to insulin. If diabetes remains untreated a lot of complications happen. But the increase of machine learning approaches solves the crucial problem.The goal of this study is to develop a model which predicts the risk of diabetes in patients with the utmost precision. In this paper we propose a prediction algorithm using modified weed optimization and the best first classifier to provide the closest results compared to clinical results.The proposed method is intended to use predictive analysis to assess the properties of diabetes miletus early detection. The performances of proposed and existing techniques are evaluated on various measures like Precision, Accuracy, F-Measure, and Recall.
Keywords: Cardiovascular disease (CVD), C-reactive protein (CRP), Intelligent Heart Disease Prediction System (IHDPS), coronary heart disease (CHD) and Improved weed optimization (IWO).