• •    

基于内容挖掘的SWIM业务数据安全研究

马兰1,王京杰2,陈焕3   

  1. 1. 中国民航大学
    2. 中国民航大学 电子信息与自动化学院 ,天津 300300
    3. 中国民航大学 电子信息与自动化学院 ,天津 300300
  • 收稿日期:2018-07-13 修回日期:2018-09-25 发布日期:2018-09-25
  • 通讯作者: 马兰

The Research on Data Security of SWIM Based on Content Mining

  • Received:2018-07-13 Revised:2018-09-25 Online:2018-09-25

摘要: 摘 要: 针对SWIM数据交换和服务共享中的数据安全问题,分析了SWIM业务流程中的安全隐患,提出了一种基于LDA主题模型的内容挖掘的恶意数据的过滤方法。首先对SWIM四种业务数据进行大数据分析,并通过LDA模型对业务数据进行特征抽取内容挖掘,最后利用KMP匹配算法在主串中查找模式串,从而检测出含有恶意关键字的SWIM业务数据。在Linux内核中对本检测方法进行测试,实验结果表明本方法能够有效的对SWIM业务数据进行内容挖掘,与其他方法相比也具有更好的检测性能。

关键词: 关键词: 内容挖掘, 关键字匹配算法, 特征匹配, 广域信息管理系统业务数据

Abstract: Abstract: In view of the data security problems in SWIM data exchange and service sharing, the security risks in the SWIM business process are analyzed, and a method of filtering malicious data based on the LDA topic model is proposed. Firstly, four kinds of SWIM business data are analyzed with big data, and the feature extraction content of business data is mined by LDA model. Finally, the pattern string is searched in the main string by using KMP matching algorithm to detect SWIM business data with malicious keywords. The test method is tested in the Linux kernel. The experimental results show that this method can effectively mine the content of SWIM business data and have better detection performance than other methods.

Key words: content mining, keyword matching algorithm, feature matching, swim business data