Output details
11 - Computer Science and Informatics
University of Kent
Improving the interpretability of classification rules discovered by an ant colony algorithm
<24> This paper proposes a new method for the discovery of unordered classification rules to improve the interpretability of the rules discovered by an Ant Colony Optimization (ACO) classification algorithm, an important problem often ignored in the literature. The paper also proposes a new measure of rule interpretability, which can be used to analyse rules discovered by any type of rule induction algorithm, not just ACO. This paper received the best in track award (out of 20 papers in the ACO and Swarm Intelligence track) of the ACM GECCO 2013 conference, the top conference in the field of evolutionary computation.