现代电子技术2025，Vol.48Issue(11)：121-130,10.DOI:10.16652/j.issn.1004-373x.2025.11.019
混合多策略北方苍鹰优化算法及特征选择
Multi-strategy hybrid northern goshawk optimization algorithm and feature selection
摘要
Abstract
In view of the slow convergence speed,low solving accuracy,and being prone to falling into local optima when the northern goshawk optimization(NGO)algorithm is used to deal with complex optimization problems,a northern goshawk optimization algorithm which integrates multiple strategies(abbreviated as LANGO)is presented in the paper.In the LANGO algorithm,the Tent chaotic mapping and opposition-based learning(OBL)strategy are used to initialize the population,increase the diversity of the population,and improve the global search ability.The nonlinear weighting factors are introduced to improve global exploration capabilities and improve the convergence speed and convergence accuracy of the algorithm.The Lévy flight is introduced to eliminate the defect of being prone to falling into local optima due to the fact that random prey is used to guide the population in the NGO algorithm.The solution that falls into local optima is disturbed to make it jump out of local optima.Eight classical benchmark functions are selected for testing,and the simulation results show that the LANGO outperforms the comparative algorithm in terms of solution accuracy,convergence speed and other aspects.The combination of LANGO and K-nearest neighbor classifier for feature selection and data classification can reduce the dimensionality of features and improve the accuracy of data classification effectively.关键词
北方苍鹰优化算法/Lévy飞行/特征选择/K近邻分类器/权重因子/收敛性Key words
NGO algorithm/Lévy flight/feature selection/K-nearest neighbor classifier/weight factor/convergence分类
电子信息工程引用本文复制引用
鲍美英,申晋祥,张景安,周建慧..混合多策略北方苍鹰优化算法及特征选择[J].现代电子技术,2025,48(11):121-130,10.基金项目
国家自然科学基金资助项目(11971277) （11971277）
山西省高等学校科技创新项目(2022L438) （2022L438）
山西大同大学科研资助项目(2020k10) （2020k10）
山西大同大学校级教学改革创新项目(XJG2023246,XJG2023251) （XJG2023246,XJG2023251）
山西省高等学校教学改革创新项目(J20241176) （J20241176）
山西大同大学云冈学专项课题(2023YGYB17) （2023YGYB17）
