2004 | OriginalPaper | Buchkapitel
TermitAnt: An Ant Clustering Algorithm Improved by Ideas from Termite Colonies
verfasst von : Vahid Sherafat, Leandro Nunes de Castro, Eduardo R. Hruschka
Erschienen in: Neural Information Processing
Verlag: Springer Berlin Heidelberg
Enthalten in: Professional Book Archive
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
This paper proposes a heuristic to improve the convergence speed of the standard ant clustering algorithm. The heuristic is based on the behavior of termites that, when building their nests, add some pheromone to the objects they carry. In this context, pheromone allows artificial ants to get more information, at the local level, about the work in progress at the global level. A sensitivity analysis of the algorithm is performed in relation to the proposed modification on a benchmark problem, leading to interesting results.