Title :
Analysis and Improvement of Extremum Random Disturbed Arithmetic Operator of a PSO Algorithm
Author :
Qingjian, Hou ; Wang Hongli
Author_Institution :
Xi´´an Res. Inst. of Hi-Tech., Hongqing, China
Abstract :
Aiming at the demerits of extremum random disturbed arithmetic operator of a particle swarm optimization algorithm, the reasonable amelioration is put forward based on the design idea of extremum random disturbed arithmetic operator. An improved particle swarm optimization algorithm is put forward and applied to parameter selection of support vector machine. The regress modeling of two common functions based on least square support vector machine is to be as examples and the simulation experiment is done. The results show that the amelioration of arithmetic operator is necessary and feasible. The convergence velocity and precision of algorithm are enhanced.
Keywords :
particle swarm optimisation; random processes; regression analysis; support vector machines; PSO algorithm; extremum random disturbed arithmetic operator; least square support vector machine; parameter selection; particle swarm optimization algorithm; regression modeling; support vector machine; Algorithm design and analysis; Arithmetic; Birds; Cities and towns; Convergence; Hybrid intelligent systems; Least squares methods; Machine learning algorithms; Particle swarm optimization; Support vector machines;
Conference_Titel :
Hybrid Intelligent Systems, 2009. HIS '09. Ninth International Conference on
Conference_Location :
Shenyang
Print_ISBN :
978-0-7695-3745-0
DOI :
10.1109/HIS.2009.89