Disease Prediction System Using Machine Learning Models

Abstract

Early identification of diseases plays a crucial role in improving healthcare outcomes and reducing treatment costs. This study presents a disease prediction system based on machine learning models that analyzes user-reported symptoms to forecast potential illnesses with high accuracy. The proposed framework collects symptom data, processes it through feature selection and encoding techniques, and applies multiple machine learning algorithms—such as Decision Tree, Random Forest, and Support Vector Machine—to determine the most probable disease. A comparative evaluation of these models identifies the most effective one in terms of accuracy, precision, and recall. Furthermore, an interactive chatbot interface is integrated to enhance user interaction by allowing individuals to input symptoms conversationally. The system aims to provide quick, reliable, and user-friendly preliminary diagnostics that can assist individuals in seeking timely medical attention. Experimental results demonstrate that the ensemble-based models outperform traditional classifiers, confirming the feasibility of machine learning techniques in healthcare prediction systems.