Detection of anomalies from user profiles generated from system logs
Corney, Malcolm W., Mohay, George M., & Clark, Andrew J. (2011) Detection of anomalies from user profiles generated from system logs. In Conferences in Research and Practice in Information Technology (CRPIT), Australian Computer Society, Inc., Curtin University, Perth, pp. 23-32.
We describe research into the identification of anomalous events and event patterns as manifested in computer system logs. Prototype software has been developed with a capability that identifies anomalous events based on usage patterns or user profiles, and alerts administrators when such events are identified. To reduce the number of false positive alerts we have investigated the use of different user profile training techniques and introduce the use of abstractions to group together applications which are related. Our results suggest that the number of false alerts that are generated is significantly reduced when a growing time window is used for user profile training and when abstraction into groups of applications is used.
Impact and interest:
Citation counts are sourced monthly from and citation databases.
Citations counts from theindexing service can be viewed at the linked Google Scholar™ search.
|Item Type:||Conference Paper|
|Keywords:||user profiling, insider misuse, abstraction|
|Subjects:||Australian and New Zealand Standard Research Classification > INFORMATION AND COMPUTING SCIENCES (080000) > COMPUTER SOFTWARE (080300) > Computer System Security (080303)|
|Divisions:||Past > Schools > Computer Science
Past > QUT Faculties & Divisions > Faculty of Science and Technology
Past > Institutes > Information Security Institute
|Copyright Owner:||Copyright 2011 [please consult the authors]|
|Deposited On:||21 Jan 2011 00:55|
|Last Modified:||23 Dec 2014 22:17|
Repository Staff Only: item control page