Support vector machine (SVM) has recently emerged as a powerful technique for solving problems in pattern classification and regression, but its performance mainly depends on the parameters selection of it. Parameters selection for SVM is very complex in nature and quite hard to solve by conventional optimization techniques, which constrains its application to some degree. PSO, as an evolutionary computing technology, has been applied successfully to various optimization problems, but has some disadvantage. So in this paper PSO is modified by added certain particles at each iterative to broaden search area, which makes particles free of local optimization. A new methodology for parameters selection of support vector regression is proposed, based on the modified PSO tuning algorithm. The methodology is used to model nonlinear dynamical system in simulation, and the simulation result assures the validity of it, not only on time but also on model accuracy
[1]
Vladimir Cherkassky,et al.
Learning from Data: Concepts, Theory, and Methods
,
1998
.
[2]
Corinna Cortes,et al.
Support-Vector Networks
,
1995,
Machine Learning.
[3]
Christian Igel,et al.
Evolutionary tuning of multiple SVM parameters
,
2005,
ESANN.
[4]
Yunqian Ma,et al.
Practical selection of SVM parameters and noise estimation for SVM regression
,
2004,
Neural Networks.
[5]
Shouyang Wang,et al.
Forecasting stock market movement direction with support vector machine
,
2005,
Comput. Oper. Res..