An Exploration of the Performance using Ensemble Methods Utilizing Random Forest Classifier for Diabetes Detection

The use of machine learning for diabetes prediction is currently a topic of active study, which is reviewed in this paper. Machine learning (ML) is a collection of methods and algorithms for extracting knowledge from data and forecasting future data. It can be used effectively to forecast results in biomedical uses like diabetes. For diabetes prognosis, supervised learning, unsupervised learning, and reinforcement learning techniques have all been used. The best strategies for diabetes prediction are identified, and the benefits and drawbacks of different research studies are discussed. Using two distinct datasets, such as the PIMA Indian diabetes dataset with 768 records and the Kaggle diabetes dataset with 952 records, the model has been assessed qualitatively and quantitatively. The proposed model obtains 0.92 f1 score, 0.96 recall score, and 0.94 precision score, representing a significant increase in prediction accuracy of 98%. The improved scores make it obvious how important the suggested work is and help the medical professionals deliver accurate results.