麻雀搜索算法改进LSSVM的网络入侵检测  

Network Intrusion Detection Based On LSSVM Improved by Sparrow Search Algorithm

作　　者：毛一鸣 程艳艳 MAO Yiming;CHENG Yanyan(School of Information Engineering,Zhengzhou University of Industrial Technology,Zhengzhou 451100,China)

机构地区：[1]郑州工业应用技术学院信息工程学院,河南郑州451100

出　　处：《太原学院学报（自然科学版）》2024年第4期64-69,共6页Journal of TaiYuan University:Natural Science Edition

基　　金：2022年教育部高等教育司产学合作协同育人资助项目(220505115250522);河南省高等教育教学改革研究与实践项目(2021SJGLX616);河南省大中专院校就业创业课题(JYB2023094)。

摘　　要：针对最小二乘支持向量机模型进行网络入侵检测的性能受其控制参数设定的影响,为提高网络入侵检测的精度,提出一种基于麻雀搜索算法优化LSSVM模型控制参数的网络入侵检测模型。与PSO-LSSVM模型、GA-LSSVM模型、GWO-LSSVM模型和LSSVM模型相比,SSA-LSSVM模型的网络入侵检测精度最高,可以实现网络入侵的高精度检测,为网络安全维护和增强入侵检测功能提供科学参考。The performance of the Least Squares Support Vector Machine model in network intrusion detection is directly affected by the penalization parameters,so a network intrusion detection model based on LSSVM model is improved by sparrow search algorithm for higher accuracy.Compared with PSO-LSSVM model,GA-LSSVM model,GWO-LSSVM model and LSSVM model,the SSA-LSSVM model has the highest accuracy of network intrusion detection.SSA-LSSVM can achieve high-precision network intrusion detection,and provide decision-making basis for network security maintenance and enhance intrusion detection function.

关 键 词：入侵检测 麻雀搜索算法 最小二乘支持向量机 召回率 精确率 

分 类 号：TP391.1[自动化与计算机技术—计算机应用技术]