摘要
由于参数的选择范围较大,在多个参数中进行盲目搜索最优参数的时间代价较大,且很难得到最优参数。为此,提出一种基于改进人工鱼群算法(AFSA)的支持向量机(SVM)预测算法。对AFSA进行改进,并使用改进算法优化SVM。实验结果表明,与遗传算法、粒子群优化算法和基本AFSA优化的支持向量机相比,该算法的均方误差降低为2.51×10 3,提高了预测精度。
For the large scale of parameter select range, it costs many time to blindly search optimal parameters in a number of parameters, and is hard to get optimal parameters. In order to solve this problem, a Support Vector Machine(SVM) prediction algorithm based on improved Artificial Fish Swarm Algorithm(AFSA) is proposed in this paper. It makes improvement with AFSA, and uses the improved AFSA to make improvement with SVM. Experimental results show that compared with Genetic Algorithm(GA), Particle Swarm Optimization(PSO) algorithm, and basic AFSA improvement SVM, the mean square error is decreased to 2.51 x 10^-3 of this algorithm, improves the prediction accuracy.
出处
《计算机工程》
CAS
CSCD
2013年第4期222-225,共4页
Computer Engineering
基金
国家自然科学基金资助项目(51275524)
关键词
支持向量机
人工鱼群算法
参数优化
回归模型
遗传算法
粒子群优化
Support Vector Machine(SVM)
Artificial Fish Swarm Algorithm(AFSA)
parameter optimization
regression model
Genetic Algorithm(GA)
Particle Swarm Optimization(PSO)