Performance Analytics of Classifiers: A Case Study with Diabetic Database

Abstract

This particular work uses artificial intelligence (AI) simulations to evaluate the accuracy of four major classifiers in diabetes prediction. Numerous efforts have been successfully made to use specific approaches to predict the negative effects of diabetes and prevent it before the disease actually manifests itself. Diabetes is extremely difficult to categorise, despite the availability of various categorization algorithms. The primary goal of this study is to compare the effectiveness of the subsequent algorithms: Random Forest (RF), Decision Tree (DT), K Nearest Neighbor (KNN), and Logistic Regression (LR) for diabetes data classification. The Pima diabetic dataset, which makes use of nine features, is made available through the UCI repository that is used in this analysis. All four algorithms' results are assessed using a range of metrics, including recall, precision, accuracy, and F-measure. The acquired results demonstrate that, in comparison to other algorithms, the RF algorithm performs with the highest accuracy of 87.01%. Receiver Operating Characteristic (ROC) curves is used to effectively and meticulously verify these data.