Title: A new hybrid metaheuristic for medical data classification

Authors: Sarab AlMuhaideb; Mohamed El Bachir Menai

Addresses: Department of Computer Science, College of Computer and Information Sciences, King Saud University, 51178 Riyadh 11543, Saudi Arabia ' Department of Computer Science, College of Computer and Information Sciences, King Saud University, 51178 Riyadh 11543, Saudi Arabia

Abstract: The classification of medical data is a complex task. Medical diagnosis and/or prognosis can be modelled as classification tasks. A hybrid metaheuristic is introduced consisting of two phases; an ant colony optimisation (ACO) phase and a genetic algorithm (GA) phase. The population of the GA is initialised to decision lists constructed during the ACO phase using different subsets of the training data. The task of the GA is to optimise the decision lists obtained in terms of classification accuracy and model size. Results on a number of benchmark real-world medical datasets show the usefulness of the proposed approach.

Keywords: medical decision support; medical data; data classification; ant colony optimisation; ACO; genetic algorithms; sequential covering; hybrid metaheuristics; medical diagnosis; medical prognosis.

DOI: 10.1504/IJMHEUR.2014.058860

International Journal of Metaheuristics, 2014 Vol.3 No.1, pp.59 - 80

Received: 06 May 2013
Accepted: 13 Sep 2013
Published online: 25 Jul 2014 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article