摘要
支持向量机的训练需要求解一个带约束的二次规划问题,但在数据规模很大的情况下,经典的训练算法将会变得非常困难。提出了一种改进的基于粒子群的优化算法,用于替代支持向量机中现有的训练算法。在改进后的粒子群优化算法中,粒子不仅向自身最优和全局最优学习,还以一定的概率向其他部分粒子的均值学习。同时,还引进了自适应变异算子,以降低未成熟收敛的概率。实验表明,提出的改进训练算法相对改进前的算法在性能上有显著提高。
Since training a SVM requires solving a constrained quadratic programming problem which becomes difficult for very large datasets,an improved particle swarm optimization algorithm is proposed as an alternative to current numeric SVM training methods.In the improved algorithm,the particles studies not only from itself and the best one but also from the mean value of some other particles.In addition,adapiive mutation is introduced to reduce the rate of premature convergence.The experimental results show that the improved algorithm is feasible and effective for SVM training.
出处
《计算机工程与应用》
CSCD
北大核心
2008年第20期138-141,共4页
Computer Engineering and Applications
基金
国家自然科学基金(the National Natural Science Foundation of China under Grant No.60472072)
陕西省自然科学基础研究计划(the Nat-ural Science Foundation of Shaanxi Province of China No.2006F05)
航空科学基金(the Aeronautical Science Foundation No.05153076)
关键词
支持向量机
粒子群优化算法
自适应变异
Support Vector Machine(SVM)
particle swarm optimization algorithm
adaptive mutation