Employing Machine Learning with Discriminative Function for Predict Heart Diseases

Heart disease is also one of the leading causes of death and therefore there is need to have a readily accurate way of diagnosing it. Based on this study, a strong basis of machine learning based on predicting heart disease has been proposed with the combination of ensemble learning algorithms (XGBo...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Journal of Al-Qadisiyah for Computer Science and Mathematics Jg. 17; H. 3
Hauptverfasser: Abdulrahman Yousif, Firas, Raad Al-Mola, Rasha, Subhi Sulaiman, Muthanna
Format: Journal Article
Sprache:Englisch
Veröffentlicht: 30.09.2025
ISSN:2074-0204, 2521-3504
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Heart disease is also one of the leading causes of death and therefore there is need to have a readily accurate way of diagnosing it. Based on this study, a strong basis of machine learning based on predicting heart disease has been proposed with the combination of ensemble learning algorithms (XGBoost, Random Forest, Gradient Boosting, and Extra Trees) and classic discriminant analysis based (Linear Discriminant Analysis (LDA) and Quadratic Discriminant Analysis (QDA)). The models were tested with two test benchmark datasets following rigorous pre-processing and feature engineering. The experimental findings show that XGBoost performed best in the first dataset with the accuracy of 98.54, perfect precision of 100 and the F1 score of 0.985. The best results by Random Forest on the second dataset, compared to the rest were 94.96 and the F1 score of 0.955. Comparatively, LDA achieved 82.44 and 87.39, accuracy rates out of the first and the second dataset respectively, whereas, QDA did not do as well in 54.15 and 44.96 respective accuracy levels. The results provided clearly indicate that compared to alternative discriminant models, machine learning models considerably outperformed the discriminant ones, which also shows that the XGBoost model would be the most suitable option in classifying the data. The outcomes confirm the usefulness of machine learning as a reliable and accurate diagnostic aid in detecting preliminary heart diseases in the clinical setting.
ISSN:2074-0204
2521-3504
DOI:10.29304/jqcsm.2025.17.32379