A Set-Based Particle Swarm Optimization Method

The representation used in Particle Swarm Optimization (PSO) is an n-dimensional vector. If you want to apply the PSO method, you have to encode your problem as fix-sized vector. But many problem domains have solutions of unknown sizes as for instance in data clustering where you often don't know the number of clusters in advance. In this paper a set-based PSO is proposed which replaces the position and velocity vectors by position and velocity sets realizing this way a PSO with variable length representation. All operations of the PSO update equations are redefined in an appropriate manner. Additionally, an operator reducing set bloating effects is introduced. The presented approach is applied to well-known data clustering problems and performs better as other algorithms on them.

[1]  Andries Petrus Engelbrecht,et al.  Fundamentals of Computational Swarm Intelligence , 2005 .

[2]  Andries Petrus Engelbrecht,et al.  Determining RNA Secondary Structure using Set-based Particle Swarm Optimization , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[3]  Zheng Qin,et al.  Research on Structure Learning of Dynamic Bayesian Networks by Particle Swarm Optimization , 2007, 2007 IEEE Symposium on Artificial Life.

[4]  Alan F. Murray,et al.  IEEE International Conference on Neural Networks , 1997 .

[5]  Bryant A. Julstrom,et al.  Edge sets: an effective evolutionary coding of spanning trees , 2003, IEEE Trans. Evol. Comput..

[6]  Bogju Lee,et al.  A set-oriented genetic algorithm and the knapsack problem , 2001, Proceedings of the 2001 Congress on Evolutionary Computation (IEEE Cat. No.01TH8546).

[7]  Yue Shi,et al.  A modified particle swarm optimizer , 1998, 1998 IEEE International Conference on Evolutionary Computation Proceedings. IEEE World Congress on Computational Intelligence (Cat. No.98TH8360).

[8]  James Kennedy,et al.  Small worlds and mega-minds: effects of neighborhood topology on particle swarm performance , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[9]  Andries Petrus Engelbrecht,et al.  Data clustering using particle swarm optimization , 2003, The 2003 Congress on Evolutionary Computation, 2003. CEC '03..

[10]  Catherine Blake,et al.  UCI Repository of machine learning databases , 1998 .

[11]  James Kennedy,et al.  Particle swarm optimization , 2002, Proceedings of ICNN'95 - International Conference on Neural Networks.

[12]  Mario Köppen,et al.  Data Swarm Clustering , 2006, Swarm Intelligence in Data Mining.

[13]  J. Kennedy,et al.  Population structure and particle swarm performance , 2002, Proceedings of the 2002 Congress on Evolutionary Computation. CEC'02 (Cat. No.02TH8600).

[14]  Marco Dorigo,et al.  On the Performance of Ant-based Clustering , 2003, HIS.