Lifestyle diseases such as diabetes, obesity, hypertension, and heart disease are increasing rapidly due to unhealthy habits, poor diet, lack of exercise, and smoking. This research presents a web-based machine learning system that predicts lifestyle disease risks using user inputs including age, gender, height, weight, exercise frequency, smoking status, diet type, family history, and medical history. The system calculates Body Mass Index (BMI) and applies multiple machine learning algorithms: Logistic Regression (82% accuracy), Decision Tree (85%), Random Forest (89%), and XGBoost (92% accuracy - best performer). Developed using Flask backend, Bootstrap frontend, and Matplotlib visualizations, the application provides real-time risk assessment (Low/Medium/High), color-coded disease-specific predictions, interactive charts, and personalized lifestyle recommendations. Unlike existing simulation-based SVM models, this system offers user-friendly web interface and preventive healthcare guidance, promoting early detection and healthy lifestyle changes.