dannydave/Heart-Disease-Diagnosis-using-Machine-Learning-and-Data-Mining ? reverse-engineered prompt

Build me a heart disease diagnosis project that uses patient clinical data to predict whether someone is at risk of heart disease.

I want the full workflow in one place, from loading the dataset and cleaning it up, to handling missing values, scaling the features, and picking the most useful inputs. Then train and compare a few models like Decision Tree, Random Forest, Naive Bayes, and SVM, and show how well each one performs with accuracy, precision, recall, F1 score, confusion matrix, and ROC AUC.

Please also include a simple Tkinter app where I can enter patient details like age, sex, chest pain type, blood pressure, cholesterol, and the rest, then get a prediction from the best model. Make it easy to run in a notebook and keep the code clear so I can understand what each step is doing. If you need to, look up current docs online for anything that might have changed.