Elkamel A.,Multimedia Information systems and Advanced Computing Laboratory Miracl |
Elkamel A.,University of Monastir |
Gzara M.,Multimedia Information systems and Advanced Computing Laboratory Miracl |
Gzara M.,University of Monastir |
Ben-Abdallah H.,King Abdulaziz University
Applied Intelligence | Year: 2014
Biological entities, such as birds with their flocking behavior, ants with their social colonies, fish with their shoaling behavior and honey bees with their complex nest construction, represent a great source of inspiration in the optimization and data mining domains. Following this line of thought, we propose the Communicating Ants for Clustering with Backtracking strategy (CACB) algorithm, which is based on a dynamic and an adaptive aggregation threshold and a backtracking strategy where artificial ants are allowed to turn back in their previous aggregation decisions. The CACB algorithm is a hierarchical clustering algorithm that generates compact dendrograms since it allows the aggregation of more than two clusters at a time. Its high performance is experimentally shown through several real benchmark data sets and a content-based image retrieval system. © 2014, Springer Science+Business Media New York.
Gzara M.,Multimedia Information systems and Advanced Computing Laboratory MIRACL |
Essabri A.,Laboratoire Of Gestion Industrielle Et Daide A La Decision Giad
Studies in Informatics and Control | Year: 2011
Most parallel evolutionary algorithms for single and multi-objective optimisation are motivated by the reduction of the computation time and the resolution of larger problems. Another promising alternative is to create new distributed schemes that improve the behaviour of the search process of such algorithms. In multi-objective optimisation problems, more exploration of the search space is required to obtain the whole or the best approximation of the Pareto Optimal Front. In this paper, we present a new clustering-based parallel multi-objective evolutionary algorithm that balances between the two main concepts in metaheuristics, which are exploration and exploitation of the search space. The proposed algorithm is implemented and tested on several standard multi-objective test functions using a network of multiple computers.
Gzara M.,Multimedia InfoRmation Systems and Advanced Computing Laboratory MIRACL |
Jamoussi S.,Multimedia InfoRmation Systems and Advanced Computing Laboratory MIRACL |
Elkamel A.,Multimedia InfoRmation Systems and Advanced Computing Laboratory MIRACL |
Ben-Abdallah H.,Multimedia InfoRmation Systems and Advanced Computing Laboratory MIRACL
Revue d'Intelligence Artificielle | Year: 2011
The behavior of real ants while resolving complex problems, that they encounter in their daily life, have inspired attracting algorithms for the resolution of the clustering problem. Most ant clustering algorithms extend the basic model developed by Lumer and Faieta (Lumer et al., 1994) which was inspired by cemetery organization and larval sorting phenomena detected on some ant species. Other algorithms were developed by taken other inspirations from real ants like self-assembling behavior and chemical recognition. In this paper we take inspiration from further biological properties of real ants mainly sensorial communication. We propose a new ant based algorithm for clustering. The basic idea of this algorithm is based on the recruitment behavior of real ants and direct communication. Another biological observation taken into account is the fact that ants can communicate with others which are far from their immediate neighborhood environment. © 2011 Lavoisier.