电子科技 ›› 2020, Vol. 33 ›› Issue (8): 80-86.doi: 10.16180/j.cnki.issn1007-7820.2020.08.014

• • 上一篇    

基于机器学习的信息安全设备调配保障技术研究

金鑫1,冯毅2,尤雪汐3,王佳欣2   

  1. 1. 中国人民解放军69260部队,新疆 乌鲁木齐 830000
    2.中国人民解放军战略支援部队信息工程大学,河南 郑州 450000
    3.中国人民解放军32124部队,吉林 延吉 133000
  • 收稿日期:2019-06-04 出版日期:2020-08-15 发布日期:2020-08-24
  • 作者简介:金鑫(1993-),男,本科。研究方向:信息安全。|冯毅(1981-),男,博士研究生,副教授。研究方向:信息安全、机器学习。
  • 基金资助:
    国家自然科学基金(61602513)

Research on Information Security Equipment Deployment Guarantee Technology

JIN Xin1,FENG Yi2,YOU Xuexi3,WANG Jiaxin2   

  1. 1. Unit 69260,PLA,Urumqi 830000,China
    2. Information Engineering University,Zhengzhou 450000, China
    3. Unit 32124,PLA,Yanji 133000,China
  • Received:2019-06-04 Online:2020-08-15 Published:2020-08-24
  • Supported by:
    National Natural Science Foundation of China(61602513)

摘要:

针对目前大量信息安全设备调配保障工作仍依靠人工方式进行调配的现状,文中提出一种基于机器学习算法的信息安全设备调配保障技术。在深入研究现有信息安全设备调配保障方案的基础上,分析方案中所有影响设备调配的属性特征;根据现有信息安全设备调配保障方案与提取的属性特征构造设备调配保障样本集;使用随机森林算法对属性特征进行优化,并选取最为合理的属性特征,结合支持向量机算法构造调配保障模型。实验对比结果表明,使用随机森林算法进行特征优化后的模型测试集准确率上升约8%,模型运行时间缩短43%,在解决实际问题过程中,与原方案的调配结果误差率下降了8.69%。

关键词: 信息安全设备, 调配保障技术, 调配方案, 随机森林, 支持向量机, 预测模型

Abstract:

AAiming at the current situation that the deployment of a large number of information security devices still relies on manual deployment, this paper proposed an information security device deployment guarantee technology based on machine learning algorithms.On the basis of in-depth study of the existing information security equipment deployment guarantee scheme, all the attribute characteristics of the equipment allocation in the scheme were analyzed. The equipment deployment guarantee sample set was constructed according to the existing information security equipment deployment guarantee scheme and the extracted attribute features. The forest algorithm optimized the attribute features and the most reasonable attribute features were selected. Combined with the support vector machine algorithm, the allocation guarantee model was constructed. The experimental comparison showed that the accuracy of the model test set with the random forest algorithm increased by nearly 8%, and the model running time was reduced by 43%. In the process of solving the actual problem, the error rate of the original plan was 8.69%.

Key words: information security equipment, deployment support technology, deployment scheme, random forest, support vector machine, predicted model

中图分类号: 

  • TP393.08