Hybridization strategies for continuous ant colony optimization and particle swarm optimization applied to data clustering
| Autor: | Hung-Yi Chang, Cheng-Lung Huang, Wen-Chen Huang, Cheng-Yi Tsai, Yi-Chun Yeh |
|---|---|
| Rok vydání: | 2013 |
| Předmět: |
Continuous optimization
Mathematical optimization Computer science Ant colony optimization algorithms MathematicsofComputing_NUMERICALANALYSIS Particle swarm optimization ComputingMethodologies_ARTIFICIALINTELLIGENCE Swarm intelligence Local optimum Pheromone Multi-swarm optimization Cluster analysis Metaheuristic Software |
| Zdroj: | Applied Soft Computing. 13:3864-3872 |
| ISSN: | 1568-4946 |
| Popis: | Ant colony optimization (ACO) and particle swarm optimization (PSO) are two popular algorithms in swarm intelligence. Recently, a continuous ACO named ACOR was developed to solve the continuous optimization problems. This study incorporated ACOR with PSO to improve the search ability, investigating four types of hybridization as follows: (1) sequence approach, (2) parallel approach, (3) sequence approach with an enlarged pheromone-particle table, and (4) global best exchange. These hybrid systems were applied to data clustering. The experimental results utilizing public UCI datasets show that the performances of the proposed hybrid systems are superior compared to those of the K-mean, standalone PSO, and standalone ACOR. Among the four strategies of hybridization, the sequence approach with the enlarged pheromone table is superior to the other approaches because the enlarged pheromone table diversifies the generation of new solutions of ACOR and PSO, which prevents traps into the local optimum. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect