A comparative evaluation of machine learning ensemble approaches for disease prediction using multiple datasets

Aug 3, 2026

Upload by Quinn

Machine learning models are applied to build and enhance disease prediction systems. Ensemble learning combines multiple classifiers to achieve more accurate results than a single model, yet a thorough comparative evaluation of ensemble variants across multiple datasets remains limited. This study uses 16 disease datasets from Kaggle and the UCI Machine Learning Repository to compare 15 ensemble technique variants. Performance is assessed with accuracy, precision, recall, F1, AUC, and AUPRC to support informed method selection for future disease prediction research.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages17

Reading Time43 min

AccessFree to download

Abstract

Introduction

Disease diagnosis and ML-based prediction

Ensemble learning rationale

Bagging, boosting, and stacking

Methods and Evaluation

Datasets and ensemble variants

Performance measures

Health and Technology (2024) 14:597–613 [https://doi.org/10.1007/s12553-024-00835-w](https://doi.org/10.1007/s12553-024-00835-w) A comparative evaluation of machine learning ensemble approaches for disease prediction using multiple datasets Palak Mahajan1 · Shahadat Uddin2 · Farshid Hajati3 · Mohammad Ali Moni4 · Ergun Gide5 Received: 16 January 2024 / Accepted: 26 February 2024 / Published online: 27 March 2024 © The Author(s) 2024, corrected publication 2024 Abstract Purpose Machine learning models are used to develop and improve various disease prediction systems. Ensemble learning is a machine learning technique that combines many classifiers to increase performance by making more accurate predictions than a single classifier. Although several researchers have employed ensemble techniques for disease prediction, a comprehensive comparative study of these techniques still needs to be provided. Methods Using 16 disease datasets from Kaggle and the UCI Machine Learning Repository, this study compares the performance of 15 variants of ensemble techniques for disease prediction. The comparison was performed using six performance measures: accuracy, precision, recall, F1 score, AUC (Area Under the receiver operating characteristics Curve) and AUPRC (Area Under the Precision-Recall Curve) . Results Stacking variant of Multi-level stacking showed superior disease prediction performance compared with other bagging and boosting variants, followed by another stacking variant (Classical stacking) . Overall, stacking outperformed bagging and boosting for disease prediction. Logit Boost showed the worst performance. Conclusion The findings ofthis study can help researchers select an appropriate ensemble approach for future studies focusing on accurate disease prediction. Keywords Bagging · Boosting · Stacking · Disease prediction · Subvariants · Performance measure Highlights: • Our research examines 15 variations of ensemble approaches for disease prediction, providing useful insights into their performance. • We evaluate the performance of these approaches using six commonly accepted performance measures: accuracy, precision, recall, F1-score, AUC (Area Under the receiver operating characteristics Curve) and AUPRC (Area Under the PrecisionRecall Curve) . • Our findings show that stacking variants, notably classical and multi-level stacking, outperform other Bagging and boosting variations in disease prediction. * Shahadat Uddin [shahadat.uddin@sydney.edu.au](shahadat.uddin@sydney.edu.au) 1 Victoria University (Sydney Campus), Sydney, Australia 2 School of Project Management, Faculty of Engineering, The University of Sydney, New South Wales, Australia 3 University of New England, Armidale, Australia 4 AI & Digital Health Technology, Artificial Intelligence and Cyber Futures Institute, Charles Sturt University, Bathurst 2795, Australia 5 Central Queensland University, Sydney, Australia 1 Introduction Disease diagnosis is a critical step in treating and managing various medical conditions. However, it can be challenging due to the complexity and variability of symptoms and signs. Correct disease diagnosis is essential for effective intervention and patient care [1] . Many scientists have developed machine learning algorithms that accurately identify a broad spectrum of diseases [2–5] . These algorithms can create disease prediction models, enabling early detection and intervention, which are crucial in reducing disease-related mortality [6] . Asa result, most medical scientists are drawn to emerging machine learning-based predictive model technologies for disease prediction. Diabetes, skin disease, kidney disease, liver disease and heart disease are all major chronic diseases that substantially impact health and, if left untreated, can lead to death [7] . Therefore, accurate disease prediction becomes vital in improving patient care and minimising the burden of these chronic conditions. By id

FAQs

What problem does the study address in disease prediction using machine learning?

It addresses the need for a comprehensive comparison of ensemble learning variants for disease prediction across multiple datasets, which is not fully covered in existing literature.

How was model performance evaluated in the study?

Performance was measured using six metrics: accuracy, precision, recall, F1 score, AUC, and AUPRC.

Which ensemble approach performed best according to the results?

The stacking variant of multi-level stacking showed the best disease prediction performance, followed by the classical stacking variant; logit boost performed worst.

Aug 3, 2026

Upload by Quinn