Heartdisease Prediction Using Machine Learning

This project studies how different health factors can influence a person’s health. It studies several attributes such asHeart Disease, High Blood Pressure, High Cholesterol, BMI, Smoking, Stroke, Diabetes, Physical Activity, DietaryHabits, Alcohol Consumption, Healthcare Accessibility and important Demographic Information (Sex, Age andEducation). Before using the data, it is checked for missing data, unusual points and analyzed using EDA to understandthe connection between different variables. New significant features are added to the data using feature engineering. Inhealthcare, Logistic Regression, Decision Trees, Random Forest, Gaussian Naive Bayes and K-Nearest Neighbours(KNN) are used to predict future health conditions. Model results are measured by using accuracy, a confusion matrixand an ROC curve. The Streamlit platform is built so that users can engage with the data, see graphs of health features,look for unusual examples and experiment with models. With this project, users gain knowledge about health data, makepredictions about health outcomes and can explore different health features.