A Prediction and Recommendation System for Diabetes Mellitus using XAI-based Lime Explainer
P. Nagaraj, V. Muneeswaran, A Dharanidharan, K Balananthanan, M. S. Arunkumar, C Rajkumar
Abstract
Diabetes Mellitus is a chronic disease and many people have been affected by this disease. Early diabetes prognosis can lead to improved treatment for the victims. Data mining methods are widely used to predict the disease early. This research [paper uses XAI (Explainable Artificial Intelligence). Various tools are used to determine the selection of key attributes and the integration, prediction, and mining rules of a diabetic organization. Since XAI is used, everyone can get the procedure without any problem. Here, four algorithms are utilized and in that, the best algorithm will be selected to obtain the best accuracy and provide a traumatic complement program. This paper has proposed a diabetes predictor model for better diabetes classification that includes a few external factors that cause diabetes as well as common factors such as Glucose, BMI, Age, Insulin, etc. Separation accuracy is improved with the new database compared to the extended database. If they do not have the precision of infection, healthy food will be provided to prevent diabetes.