ENHANCING DIABETES PREDICTION USING SUPERVISED MACHINE LEARNING: A COMPARATIVE ANALYSIS OF RANDOM FOREST AND LOGISTIC REGRESSION WITH FEATURE SELECTION

ABSTRACT

Diabetes remains a significant global health challenge, affecting millions and leading to severe complications if not detected early. Accurate and timely diagnosis is crucial for effective management and intervention. This study investigates diabetes prediction using supervised machine learning, comparing the performance of Logistic Regression and Random Forest models, both with and without feature selection. Theobjective is to assess the impact of feature selection on model accuracy, precision, recall, and overall classification performance. Utilizing the Pima Indians Diabetes Database, the research evaluates the effectiveness of these models in handling medical data for early diabetes detection. The findings indicate that Random Forest consistently outperforms Logistic Regression, demonstrating superior predictive capability. This study contributes to the advancement of machine learning application sin healthcare, enhancing the reliability of diabetes diagnosis and supporting data-driven clinical decision-making.