An Efficient Machine Learning Classification Model for Diabetes Prediction

Abstract

Chronic diseases include diabetes is one of the deadliest diseases that still poses a severe world-wide health threat because it has an impact on everyone’s health. It’s a genetic condition. that results in excess glucose levels and a number of other issues, including stroke, renal failure, heart and nerve issues, as well as high glucose levels. Numerous researchers have worked to develop a reliable diabetes prediction model throughout the years. Because there are still many unanswered research questions in this field and there aren’t enough reliable data sets and forecasting tools, researchers are turning to deep learning and ML-based approaches. Three different ML algorithms are used in the study to explore issues as it looks at healthcare predictive analytic. The primary purpose of this study was to look at the possible uses of deep learning and machine learning techniques for diabetes therapy. The outcomes indicate the possibility of a high level of accuracy for the machine learning-based architecture. Stakeholders and healthcare professionals are creating categorization models to help with diabetes diagnosis and the creation of prevention strategies. Based on their results, they conduct a literature review of machine model research and offer a unique paradigm for diabetes prediction. Before developing and putting to the test an intellectual ML-based architecture for mellitus forecasting, the researchers conduct a critical analysis of machine learning models. When compared to various basic classifiers for diabetes prediction, such as the DT, SVM, and EDPA (Ensemble Diabetic Prediction Algorithm) ML approaches, the suggested classifier outperformed when compared with the other 2 ML algorithms.

Topics & Concepts