원문정보
초록
영어
A clustering algorithm is proposed in this paper, which is based on discussion of multi-agent meta-heuristic architecture of the ant colony optimization algorithm. The multi-agent architecture of ant colony optimization meta-heuristic includes three levels. Level-0 agents build solutions, level-l agents improve solutions and level-2 agents update pheromone matrix. The updated pheromone then provides feedback information for the next iteration of solution construction. Mutation probability p and pheromone resistance ρ are the adaptive parameters, which can be adjusted automatically during the evolution progress. With the adaptive variable, the algorithm can solve the contradiction between convergence speed and precocity and stagnation. The algorithm has been tested and compared with the clustering algorithm based on Genetic and Simulate annealing. Experimental results show that the proposed algorithm is more effective, and the clustering quality and efficiency are promising.
목차
1. Introduction
2. Multi-Agent Architecture of ACO Meta-heuristic
3. Adaptive ACO Clustering Algorithm
3.1. Definition of clustering problem
3.2. Coding and criterion function
3.3. Solution construction of Level 0
3.4. Local search of Level 1
3.5. Pheromone updating with adaptive mechanism of Level 2
4. Experimental Results
5. Conclusion
Acknowledgements
References