网络与信息安全学报 ›› 2018, Vol. 4 ›› Issue (1): 45-51.doi: 10.11959/j.issn.2096-109x.2018011

• 学术论文 • 上一篇    下一篇

基于机器学习的Web用户行为认证

毋泽南1,田立勤1,2(),王志刚2   

  1. 1 华北科技学院计算机学院,北京 101601
    2 青海师范大学计算机学院,青海 西宁810008
  • 修回日期:2018-01-05 出版日期:2018-01-01 发布日期:2018-02-09
  • 作者简介:毋泽南(1991-),男,河南焦作人,华北科技学院硕士生,主要研究方向为信息安全、用户行为认证。|田立勤(1970-),男,陕西定边人,博士,华北科技学院教授,主要研究方向为物联网远程信息监控、大数据有效性审核、网络安全评价与用户行为认证、网络性能评价与优化。|王志刚(1993-),男,河北张家口人,青海师范大学硕士生,主要研究方向为大数据、云计算。
  • 基金资助:
    国家自然科学基金资助项目(61472137);中央高校基本科研业务费基金资助项目(3142015022);中央高校基本科研业务费基金资助项目(3142017053);青海省重点研发、应用基础研究基金资助项目(2016-SF-130);青海省重点研发、应用基础研究基金资助项目(2017-ZJ-752);河北省物联网监控工程技术研究中心基金资助项目(3142016020);青海省物联网重点实验室基金资助项目(2017-ZJ-Y21)

Behavior authentication of Web users based on machine learning

Zenan WU1,Liqin TIAN1,2(),Zhigang WANG2   

  1. 1 School of Computer Science and Technology,North China Institute of Science and Technology,Beijing 101601,China
    2 School of Computer Science and Technology,Qinghai Normal University,Xining 810008,China
  • Revised:2018-01-05 Online:2018-01-01 Published:2018-02-09
  • Supported by:
    The National Natural Science Foundation of China(61472137);The Central University Basic Research Fees Fund Project(3142015022);The Central University Basic Research Fees Fund Project(3142017053);The Key Research and Development Projects of Qinghai Province(2016-SF-130);The Key Research and Development Projects of Qinghai Province(2017-ZJ-752);Internet of Things Monitoring Engineering Research Center of Hebei Province(3142016020);Key Laboratory of Internet of Things of Qinghai Province(2017-ZJ-Y21)

摘要:

针对Web用户信息的安全问题,结合机器学习的方法,对用户行为进行分析和认证。首先通过主成分分析法对原始数据集做降维处理,然后利用 SVM 算法,让计算机对历史用户行为证据进行学习,得到一个判别用户身份的模型。实际应用和理论分析表明,该模型在用户行为认证判别上,可以更准确和高效地分类出危险用户和可信用户,为诸如电子商务、网络金融等关键网络应用用户行为的高性能分析奠定坚实的理论和实践基础。

关键词: 信息安全, 用户行为认证, 支持向量机, 主成分分析

Abstract:

According to the security problem of Web user information,the user behavior was analyzed and authenticated by the method of machine learning.First of all,through the principal component analysis to reduce the dimension of the original data set,then use the SVM algorithm to allow the computer to learn the history of user behavior evidence,to get a model to identify the user's identity.The practical application and theoretical analysis show that the model in user behavior identification authentication,can be more accurate and efficient classification of dangerous users and trusted users,analysis lay a solid theoretical and practical basis for the high performance user behavior such as electronic commerce,network finance and other key of Internet applications.

Key words: information security, user behavior authentication, support vector machine, principal component analysis

中图分类号: 

No Suggested Reading articles found!