Outlier detection in audit logs for application systems

作者：

Highlights：

• Evaluate a single algorithm is not effective for the generalization of the problem that the detection means fields considered outliers.

• Combined different algorithms to optimize the detection of outliers are evaluated.

• Designed a processes that combine Data Mining algorithms to detect outliers fields.

• The convenience of applying hybrid methods in the detection of outliers is evaluated.

摘要

•Evaluate a single algorithm is not effective for the generalization of the problem that the detection means fields considered outliers.•Combined different algorithms to optimize the detection of outliers are evaluated.•Designed a processes that combine Data Mining algorithms to detect outliers fields.•The convenience of applying hybrid methods in the detection of outliers is evaluated.

论文关键词：Data mining,Systems audit,Outlier detection

论文评审过程：Received 15 January 2014, Revised 5 March 2014, Accepted 8 March 2014, Available online 26 March 2014.

论文官网地址：https://doi.org/10.1016/j.is.2014.03.001