Prediction of Diabetic Disease through Big Data Techniques: A Comparative Study

Abstract

Medical data mining is the method of collecting, examining and interpreting the health information for improving the patient care and support decision-making. Medical data processing comprised the data collection, storage, analysis and prediction. Diabetes is a chronic condition characterized by impaired glucose metabolism. Diabetes poses the significant global health challenge. Diabetes resulted in chronic damage and dysfunction of different tissues like eyes, kidneys, heart, blood vessels and nerves. Diabetes prediction has received large research interest to emphasize the importance of predicting diabetes for early intervention and management strategies. Data pre-processing is the process of handling the missing the data points from input dataset. Feature selection is the process of selecting the relevant features from input database. With the selected features, the patient data classification is carried out. Different researchers carried out their research on different diabetes disease detection. But, the accuracy level was not enhanced and time complexity was not reduced. In order to address these issues, machine learning and deep learning based diabetic disease diagnosis is discussed in the survey article.