摘要
为避免大规模信息入侵行为的出现,提出基于机器学习的舰船信息系统入侵检测技术。基于机器学习原理分析舰船信息系统的具体组成形式,根据入侵数据挖掘标准计算信息相似度指标与检测修正系数,实现舰船信息系统入侵检测算法的设计与应用。实例分析结果表明,若同时存在多种丢弃模式,则机器学习算法作用下的舰船信息系统数据会话延迟时间始终略低于理想时长,能够较好抑制大规模信息入侵行为的出现。
In order to avoid the emergence of large-scale information intrusion behaviors,a ship information system intrusion detection technology based on machine learning is proposed.Based on the machine learning principle,the specific composition of the ship information system is analyzed,and the information similarity index and detection correction coefficient are calculated according to the intrusion data mining standard to realize the design and application of the ship information system intrusion detection algorithm.The case analysis results show that if there are multiple discarding modes at the same time,the ship information system data session delay time under the action of the machine learning algorithm is always slightly lower than the ideal time,which can better suppress the appearance of large-scale information intrusion.
作者
马海洲
丁爱萍
MA Hai-zhou;DING Ai-ping(Yellow River Conservancy Technical Institute,College of Information Engineering,Kaifeng 475001,China)
出处
《舰船科学技术》
北大核心
2021年第22期163-165,共3页
Ship Science and Technology
基金
河南省高等教育教学改革研究与实践项目(2017SJGLX140)
关键词
机器学习
信息系统
入侵检测
数据挖掘
信息相似度
修正系数
machine learning
information system
intrusion detection
data mining
information similarity
correction coefficient
