Title :
Support vector machine with parameter optimization by bare bones differential evolution
Author :
Daoyin Qiu ; Yao Li ; Xiaoyuan Zhang ; Bo Gu
Author_Institution :
Huazhong Univ. of Sci. & Technol., Wuhan, China
Abstract :
As one state-of-the-art pattern recognition method, support vector machine (SVM) has been successfully applied in many diverse regions. But the parameter optimization for SVM is a further ongoing research issue. The most used grid search method is time-consuming and the traditional global stochastic optimization techniques are parameter dependent. In this paper, bare bones differential evolution (BBDE) is used to tune the parameters of SVM. The BBDE is a new, almost parameter-free optimization algorithm that is a hybrid of the barebones particle swarm optimization (PSO) and differential evolution (DE). Some international standard data sets are used to evaluate the proposed algorithm. The experiment shows that BBDE has good performances to find the optimal parameters for SVM and is superior to some other methods.
Keywords :
evolutionary computation; particle swarm optimisation; pattern recognition; search problems; stochastic processes; support vector machines; bare bones differential evolution; global stochastic optimization techniques; grid search method; parameter-free optimization algorithm; particle swarm optimization; pattern recognition method; support vector machine; Algorithm design and analysis; Bones; Kernel; Optimization; Particle swarm optimization; Support vector machines; Testing; Support vector machine; bare bones differential evolution; differential evolution; particle swarm optimization;
Conference_Titel :
Natural Computation (ICNC), 2011 Seventh International Conference on
Conference_Location :
Shanghai
Print_ISBN :
978-1-4244-9950-2
DOI :
10.1109/ICNC.2011.6022065