Hybrid and Effective Model for the Prediction of Diabetes

Abstract The most widely studied research area in healthcare is healthcare systems using modern integrated computing techniques. A lot of data is generated from innu- merable heterogeneous healthcare sensors, IoT devices, and monitoring devices. Collecting, organizing, understanding, and forecasting patient health is extremely important. In this research paper, a smart healthcare recommendation system, namely, Hybrid and Effective Prediction of Diabetes (HEPD), is proposed. HEPD uses data fusion techniques and machine learning methods to predict and recom- mend treatment for diabetes and other life-threatening diseases more accurately. It is an intelligent recommendation system that is trained to predict diabetes. For in-depth evaluation of this HEPD model, it is simulated and examined on estab- lished heterogeneous datasets. The outcome of the simulations is analogized with the most recent development and existing models. From the comparison results, it is found that the HEPD achieves 91.5% accuracy, which is much higher than the renowned machine learning methods.