PHA-ML: Prediction heart attack using machine learning algorithms and apache spark

A heart attack is considered a disease that affects human life and may cause sudden death. Currently, many researchers and scientists are seeking to find solutions and predict the occurrence to save as many lives as possible. This paper proposes a prediction system to predict heart attack disease from Kaggle data (healthcare), powered by Spark 2.3.1. The aim is to find the most accurate machine-learning model for predicting heart attack disease and disaggregating the results. In predicting heart attack disease, the full features are validated by evaluating and comparing them with the machine learning models. In this research, a comparative study with learning algorithms was conducted, including feature selection, feature selection focusing on RFE, Feature selection using univariate cross-validation, and classification. The results of the techniques, Logistic regression, Decision Tree Classifier, Linear SVC, and Random Forest Classifier, have been compared. The results showed that the Random Forest Classifier is the most effective and accurate prediction method among the other techniques used, with 88% accuracy at feature selection using full-feature-cross-validation, 71% at feature selection by RFE, and 88% at feature selection by univariate cross-validation. Therefore, this paper focuses on the random forest classifier used to predict heart attack disease. It is a comprehensive and effective classification model in machine learning. Results confirm the ability to predict heart disease and help improve health outcomes, providing accurate diagnoses, which greatly contribute to patient care.