摘要
针对网络入侵的特征,提出一种基于SVM支持向量机的入侵危险识别模型。利用支持向量机SVM模型,混合人工蜂群HABC优化的方式,克服算法中存在早熟收敛和局部极小的问题。通过该模型实现对网络入侵信息系统自适应识别出攻击效果,有效得到网络入侵的信息系统风险评估。验证结果表明,HABC优化的SVM模型比传统危险入侵识别模型的准确度更高,收敛速度快,泛化能力增强,说明了该方法的可行性、有效性。
Aiming at the characteristics of the network intrusion, an intrusion risk identification model based on support vector machine (SVM) is proposed. The SVM model and the optimization method of the hybrid artificial bee colony (HABC) are used to overcome the problems of premature convergence and local minimum existing in the algorithm. The attack effect of the network intrusion information system can be recognized automatically with this model, and the information system risk assess- ment of network intrusion can be obtained effectively. The experimental results show that the SVM model optimized with HABC has higher accuracy than that of the traditional danger intrusion identification models, faster convergence rate and stronger genera- lization ability, and the feasibility and effectiveness of the proposed method are verified.
出处
《现代电子技术》
北大核心
2017年第7期81-84,共4页
Modern Electronics Technique
关键词
安全监测
混合人工蜂群算法
支持向量机
信息安全
风险评估
safety monitoring
hybrid artificial bee colony algorithm
support vector machine
information security
riskassessment