摘要
为了准确识别物联网环境下舰船监控网络高维异常数据,针对当前识别方法存在的误差大、速度慢等不足,提出一种物联网环境下舰船监控网络高维异常数据挖掘方法。首先分析当前物联网环境下舰船监控网络高维异常数据识别的研究现状,指出各种方法的局限性,然后结合舰船监控网络异常数据的高维特点,引入解决了"维数灾"问题的支持向量机对舰船监控网络高维异常数据进行挖掘,找到舰船监控网络异常数据的变化趋势,最后通过仿真实验分析了其有效性和优越性。结果表明,本文方法提高了舰船监控网络高维异常数据识别正确率,误识率明显下降,减少了舰船监控网络高维异常数据识别时间,可以对大规模舰船监控网络高维异常数据进行处理,具有广泛的应用前景。
In order to accurately identify the high-dimensional abnormal data of warship monitoring network in the Internet of Things,a data mining method of high-dimensional abnormal data of warship monitoring network in the Internet of Things is proposed to solve shortcomings of current identification methods,such as errors and slow speed.Firstly,the current research status of high-dimensional abnormal data recognition in ship monitoring network under the Internet of Things is analyzed,Then,combined with high-dimensional characteristics of abnormal data in ship monitoring network,support vector machine is introduced to solve problem of"dimension disaster"to mine high-dimensional abnormal data in ship monitoring network.Finally,the effectiveness and superiority of the proposed method are analyzed by simulation experiments.The results show that the proposed method can improve recognition accuracy of high-dimensional anomalous data in ship monitoring network,and significantly reduce recognition time of high-dimensional anomalous data in ship monitoring network.It can process high-dimensional anomalous data in large-scale ship monitoring network and has a wide application prospect.
作者
李洪波
LI Hong-bo(School of Information and Electrical Engineering,Ludong University,Yantai 264025,China)
出处
《舰船科学技术》
北大核心
2019年第20期154-156,共3页
Ship Science and Technology
基金
山东省自然科学基金面上项目(ZR2010GM013).
关键词
物联网
舰船监控网络
异常数据
高维特性
internet of things
ship monitoring network
abnormal data
high-dimensional characteristics
