Authors: Samruddhi Ande, Professor S. V. Raut
Abstract: Millions of individuals worldwide suffer with diabetes, a dangerous medical condition. Serious problems can be avoided with early diabetes prediction. In this study, we predict diabetes in individuals based on a variety of health factors using machine learning approaches. Age, blood pressure, glucose level, BMI, and other medical characteristics are among the data in the dataset. To increase prediction accuracy, data preprocessing techniques such as normalization and handling missing values were used. A number of machine learning models were tested, such as Support Vector Machine, Random Forest, and Decision Tree. The accuracy, precision, recall, and F1-score of these models were used to compare their performances. The Random Forest model demonstrated its suitability for diabetes prediction by achieving the best accuracy. The findings show that machine learning may reliably support early diagnosis, assisting physicians and patients in making better health-related decisions. The significance of technology in healthcare and the potential for AI-based solutions to enhance patient outcomes are highlighted in this study.
