论文部分内容阅读
针对蚁群算法存在易过早收敛、出现停滞现象、陷入局部极值的问题,提出S型信息素更新策略与Alopex算法相耦合的改进蚁群优化算法(IACO).该算法定义全新的S型动态自适应信息素全局更新函数,使信息素增量随迭代次数和目标函数值变化而动态变化,同时耦合Alopex算法以提高算法的局部搜索能力.将IACO算法应用于支持向量机参数的优化中,构成IACO-SVM模型.利用UCI标准数据集进行数值实验.研究结果表明:IACO算法具有较强的寻优性能,IACO-SVM模型具有较高的平均分类准确率和较好的稳定性.
Aiming at the problem of premature convergence, stagnation and falling into local extremum in ant colony algorithm, an improved ant colony optimization algorithm (IACO) is proposed, which combines the S-pheromone updating strategy and Alopex algorithm.The algorithm defines a new S-type Dynamically update the pheromone function dynamically so that the increments of pheromones vary dynamically with the number of iterations and the value of objective function.Alopex algorithm is also used to improve the local search ability of the algorithm.The IACO algorithm is applied to optimize the parameters of SVM , And the IACO-SVM model is constructed.Using the UCI standard dataset, numerical experiments are carried out.The results show that the IACO algorithm has better performance in optimization and the IACO-SVM model has higher average classification accuracy and better stability.