摘要
机器学习的性能受特征选择和参数优化的影响很大,针对这一问题,采用基于蚁群算法和遗传算法的混合算法对特征选择和参数优化问题进行了探究。实验结果表明,该混合算法相比单个的蚁群算法或遗传算法,在特征选择和参数优化方面,具有更高的准确率。
The performance of machine learning is greatly influenced by feature selection and parameter optimization.To address this issue,a hybrid algorithm based on ant colony and genetic algorithms was explored for feature selection and parameter optimization.Experimental results show that this hybrid algorithm has higher accuracy in feature selection and parameter optimization compared to individual ant colony or genetic algorithms.
作者
李中民
Li Zhongmin(Party School of the Guangzhou Committee of C.P.C,Guangzhou 510070,China)
出处
《现代计算机》
2023年第13期50-54,共5页
Modern Computer
关键词
机器学习
特征选择
参数优化
蚁群算法
遗传算法
machine learning
feature selection
parameter optimization
ant colony algorithm
genetic algorithm
