A Proposed Tree-based Explainable Artificial Intelligence Approach For The Prediction Of Angina Pectoris

This paper presents a new dataset for AP disease and compares different Machine Learning algorithms to predict AP. An explainable approach based on Generalized Additive Models (GAMs) was developed to accurately predict AP events with the interpretation of the results. This is the first study to use Explainable AI/ML (XAI/XML) to predict AP events. The dataset used in the study was obtained from the web address (https://www.kaggle.com/datasets/snehal1409/predict-angina) and included 200 female patients who were examined for the presence or absence of AP, as well as for several other variables that could potentially influence the development of AP. The variables that were examined for each patient were smoking habits, age, family history of angina, hypertension status, amount of cigarettes consumed per day, family history of myocardial infarction, and family history of stroke and diabetes.