Historical behavior baseline modeling and anomaly detection in machine generated end to end event log

作者: Derek Lin , Chunsheng Fang

DOI:

关键词:

摘要: A bipartite graph is generated which includes one or more source vertices and destination vertices. For a given vertex, temporal behavioral matrix using the where first dimension of associated with time second at least some model portion matrix. Anomaly detection performed on part model.

参考文章(12)
Nong Ye, A Markov Chain Model of Temporal Behavior for Anomaly Detection information assurance and security. ,(2000)
Alexander Beutel, Wanhong Xu, Detection of lockstep behavior ,(2013)
Walter Dean Mieher, Wen Jin, Vi Vuong, Determination of training set size for a machine learning system ,(2010)
Konstantin S. Ignatyev, Alexander V. Ledenev, Evgeny B. Kolotinsky, System and method for adaptive control of user actions based on user's behavior ,(2013)
Wei Wang, Xiaohong Guan, Xiangliang Zhang, Profiling program and user behaviors for anomaly intrusion detection based on non-negative matrix factorization conference on decision and control. ,vol. 1, pp. 99- 104 ,(2004) , 10.1109/CDC.2004.1428613
Staffan Klensmeden, Radislav Alexandrovich Potyrailo, Vincent Francis Pizzi, Richard John Ferraro, System and method for integrating rfid sensors in manufacturing system comprising single use components ,(2008)