HEALTH CONDITION PREDICTOR FOR LIFESTYLE RISKS

Abstract

Lifestyle diseases—such as diabetes, cardiovascular disorders, hypertension, and obesity—are primarily associated with the behavioral and environmental choices made by individuals or communities. Despite the availability of vast amounts of disease-related data generated within the healthcare sector, this information is rarely mined to uncover hidden patterns or predictive insights that could significantly improve clinical decision-making and preventive care.The objective of this study is to investigate the effectiveness of Support Vector Machines (SVM), a widely used supervised learning algorithm, in predicting an individual's susceptibility to lifestyle-related diseases. Beyond prediction, the study proposes an economically viable machine learning–based diagnostic framework that may serve as an alternative to costly deoxyribonucleic acid (DNA) testing. By analyzing lifestyle attributes—such as diet quality, caloric consumption, physical activity levels, sleep patterns, and stress factors—the model identifies potential disease risks formed by chronic unhealthy habits.The simulated model acts as an intelligent, low-cost solution capable of detecting early indicators of disorders that may mimic or lead to genetic-level abnormalities due to prolonged lifestyle imbalances. This approach lays a foundation for preventive healthcare, enabling early intervention, risk assessment, and enhanced disease management without the financial burden of conventional DNA-based diagnostic techniques.