Output details
11 - Computer Science and Informatics
University of Kent
A hierarchical multi-label classification ant colony algorithm for protein function prediction
<24> This paper proposes the first ant colony classification algorithm for hierarchical multi-label classification, a much more complex case of classification than the conventional (flat) classification. This is applied to prediction of protein function, a significant and challenging bioinformatics problem. The proposed algorithm was rigorously compared with state-of-the-art algorithms for hierarchical multi-label classification and achieved excellent results. Other authors have subsequently used the proposed algorithm in performance comparisons for hierarchical multi-label classification problems. The paper was published in the special issue on “metaheuristics for large scale data mining” of the Memetic Computing journal (acceptance rate for this issue: 22%).