Vol.27, No.3, 2021, pp.683-699, doi:10.32604/iasc.2021.014811
A Multi-Agent Stacking Ensemble Hybridized with Vaguely Quantified Rough Set for Medical Diagnosis
  • Ali M. Aseere1,*, Ayodele Lasisi2
1 Department and College of Computer Science, King Khalid University, Abha, Kingdom of Saudi Arabia
2 Department of Mathematical Sciences, Faculty of Science, Augustine University, Ilara-Epe, Lagos, Nigeria
* Corresponding Author: Ali M. Aseere. Email:
Received 19 October 2020; Accepted 20 December 2020; Issue published 01 March 2021
In the absence of fast and adequate measures to combat them, life-threatening diseases are catastrophic to human health. Computational intelligent algorithms characterized by their adaptability, robustness, diversity, and recognition abilities allow for the diagnosis of medical diseases. This enhances the decision-making process of physicians. The objective is to predict and classify diseases accurately. In this paper, we proposed a multi-agent stacked ensemble classifier based on a vaguely quantified rough set, simple logistic algorithm, sequential minimal optimization (SMO), and JRip. The vaguely quantified rough set (VQRS) is used for feature selection and eradicating noise in the data. There are two classifier layers in the stacked ensemble, with the simple logistic and SMO being the base classifiers, and JRip being the meta-classifier. The base classifiers learn using training data, and the individual classifier predictions are merged for input into the meta-classifier, which makes predictions accordingly. In experiments, the proposed method was validated by providing improved diagnoses with respect to three classification operations. The proposed method thus provides more accurate medical diagnoses to the compared algorithms.
Classification; feature selection; medical diagnosis; predictive analysis; stacking; vaguely quantified rough set
Cite This Article
A. M. Aseere and A. Lasisi, "A multi-agent stacking ensemble hybridized with vaguely quantified rough set for medical diagnosis," Intelligent Automation & Soft Computing, vol. 27, no.3, pp. 683–699, 2021.
This work is licensed under a Creative Commons Attribution 4.0 International License , which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.