摘要
针对传统的聚类算法只能满足一般高维数据的实时动态挖掘问题,提出一种基于选择分形聚类算法的用户异常行为检测算法,通过提高分形聚类准确性和有效性,同时可以实现数据的挖掘。测试结果表明,该算法的检测率和准确率达92%以上,误报率低于1.5%,具有良好的鲁棒性,可以有效地检测和识别云计算用户的异常行为,帮助云服务提供商及时发现和解决问题,提高云服务的可靠性和安全性。
With more and more resources of mobile cloud computing shifting from local to cloud,various hacker and intrusion behaviors emerge endlessly,and the attack methods are diverse.Traditional passive defense methods are difficult to effectively solve the security issues of mobile cloud computing users.The best way to analyze abnormal users’behavior is to gather normal behavior data together and cluster abnormal behavior data together.Aiming at the problem that traditional clustering algorithms can only meet the real-time dynamic mining of general high-dimensional data,a users’abnormal behavior detection algorithm based on the selection fractal clustering algorithm is proposed.By improving the accuracy and effectiveness of fractal clustering,data mining can also be achieved.The test results show that the detection rate and accuracy of the algorithm are over 92%,with a false alarm rate of less than 1.5%.It has good robustness and can effectively detect and identify abnormal behavior of cloud computing users,helping cloud service providers to discover and solve problems in a timely manner,and improving the reliability and security of cloud services.
作者
余建
林志兴
肖香梅
YU Jian;LIN Zhixing;XIAO Xiangmei(Network Technology Center,Sanming University,Sanming 365004,China;School of Information Engineering,Sanming University,Sanming 365004,China)
出处
《三明学院学报》
2025年第3期27-38,74,共13页
Journal of Sanming University
基金
福建省高校教育信息化课题(FJGX2023013)
福建省教育厅中青年课题(JAT230720)
福建省财政厅科技专项课题(KB25007)。
关键词
云计算
分形维数
聚类
异常行为检测
cloud computing
fractal dimension
clustering
abnormal behavior detection
