基于改进智能优化算法的数据特征选择方法  

Data feature selection method based on improved intelligent optimization algorithm

作　　者：张玉琴[1] 张建亮[1] 冯向东[1] ZHANG Yuqin;ZHANG Jianliang;FENG Xiangdong(The Engineering&Technical College of Chengdu University of Technology,Leshan 614000,China)

机构地区：[1]成都理工大学工程技术学院,四川乐山614000

出　　处：《传感器与微系统》2023年第6期154-157,共4页Transducer and Microsystem Technologies

基　　金：四川省教育厅自然科学重点项目(18ZA0075,18ZA0073);成都理工大学工程技术学院基金资助项目(C122019027)。

摘　　要：特征选择在人工智能领域是一个困难又重要的问题,为了快速准确地筛选特征,提出一种改进的帝王蝶优化(IMBO)算法进行封装特征选择。首先,为了使新产生的个体具有一定种群记忆功能,在迁徙算子中添加5种差分变异策略,提高种群个体的信息共享。然后,封装式特征选择算法使用模糊C均值(FCM)分类器作为评估器,利用IMBO作为搜索策略。实验结果表明:与原帝王蝶优化(MBO)、人工蜂群(ABC)算法和蚱蜢优化算法(GOA)相比,所提算法在多个公开数据集上实现了较高的准确度,显著降低了特征选择规模,且具有较快的收敛速度。Feature selection is a difficult and important problem in the field of artificial intelligence(AI).To screen features quickly and accurately,an improved monarch butterfly optimization(IMBO)algorithm is proposed to select encapsulated features.Firstly,in order to make the new individuals have a certain population memory function,five differential mutation strategies are added to the migration operator to improve the information sharing of individual of population.Then,the encapsulated feature selection algorithm uses fuzzy C-means(FCM)classifier as evaluator and IMBO as search strategy.Experimental results show that,compared with the original MBO,artificial bee colony(ABC)and grasshopper optimization algorithm(GOA),the proposed algorithm achieves higher accuracy on multiple public datasets,significantly reduces the scale of feature selection,and has faster convergence speed.

关 键 词：帝王蝶优化 模糊C均值 特征选择 收敛速度 分类 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]