Machine Learning Project: Heart Disease Classification

• This is an end-to-end machine learning project that aims to predict whether or not a patient has heart disease based on his/her medical attributes.

• Matplotlib was used for data visualization and Scikit-Learn's estimators for modeling.

• RandomizedSearchCV and GridSearchCV were used to tune hyperparameters.

• Scikit-Learn's LogisticRegression estimator produced the highest cross-validated accuracy score of 83% compared to KNeighborsClassifier and RandomForestClassifier.

• The features with the most impact to the model are sex, exercise induced angina (exang), chest pain type (cp), thalassemia (thal), and number of major vessels (0-3) colored by flourosopy (ca).

• Check out the project here.