An Attention-Based GRU Network for Anomaly Detection from System Logs

Yixi XIE; Lixin JI; Xiaotao CHENG

doi:10.1587/transinf.2020EDL8016

An Attention-Based GRU Network for Anomaly Detection from System Logs

Yixi XIE, Lixin JI, Xiaotao CHENG

Full Text Views

0

Cite this

Summary :

System logs record system states and significant events at various critical points to help debug performance issues and failures. Therefore, the rapid and accurate detection of the system log is crucial to the security and stability of the system. In this paper, proposed is a novel attention-based neural network model, which would learn log patterns from normal execution. Concretely, our model adopts a GRU module with attention mechanism to extract the comprehensive and intricate correlations and patterns embedded in a sequence of log entries. Experimental results demonstrate that our proposed approach is effective and achieve better performance than conventional methods.

Publication: IEICE TRANSACTIONS on Information Vol.E103-D No.8 pp.1916-1919

Publication Date: 2020/08/01

Publicized: 2020/05/01

Online ISSN: 1745-1361

DOI: 10.1587/transinf.2020EDL8016

Type of Manuscript: LETTER

Category: Information Network

Authors

Yixi XIE
  Information Engineering University
Lixin JI
  Information Engineering University
Xiaotao CHENG
  Information Engineering University

Keyword

anomaly detection, GRU, attention-based model

Cite this

Copy

Yixi XIE, Lixin JI, Xiaotao CHENG, "An Attention-Based GRU Network for Anomaly Detection from System Logs" in IEICE TRANSACTIONS on Information, vol. E103-D, no. 8, pp. 1916-1919, August 2020, doi: 10.1587/transinf.2020EDL8016.
Abstract: System logs record system states and significant events at various critical points to help debug performance issues and failures. Therefore, the rapid and accurate detection of the system log is crucial to the security and stability of the system. In this paper, proposed is a novel attention-based neural network model, which would learn log patterns from normal execution. Concretely, our model adopts a GRU module with attention mechanism to extract the comprehensive and intricate correlations and patterns embedded in a sequence of log entries. Experimental results demonstrate that our proposed approach is effective and achieve better performance than conventional methods.
URL: https://global.ieice.org/en_transactions/information/10.1587/transinf.2020EDL8016/_p

Copy

@ARTICLE{e103-d_8_1916,
author={Yixi XIE, Lixin JI, Xiaotao CHENG, },
journal={IEICE TRANSACTIONS on Information},
title={An Attention-Based GRU Network for Anomaly Detection from System Logs},
year={2020},
volume={E103-D},
number={8},
pages={1916-1919},
abstract={System logs record system states and significant events at various critical points to help debug performance issues and failures. Therefore, the rapid and accurate detection of the system log is crucial to the security and stability of the system. In this paper, proposed is a novel attention-based neural network model, which would learn log patterns from normal execution. Concretely, our model adopts a GRU module with attention mechanism to extract the comprehensive and intricate correlations and patterns embedded in a sequence of log entries. Experimental results demonstrate that our proposed approach is effective and achieve better performance than conventional methods.},
keywords={},
doi={10.1587/transinf.2020EDL8016},
ISSN={1745-1361},
month={August},}

Copy

TY - JOUR
TI - An Attention-Based GRU Network for Anomaly Detection from System Logs
T2 - IEICE TRANSACTIONS on Information
SP - 1916
EP - 1919
AU - Yixi XIE
AU - Lixin JI
AU - Xiaotao CHENG
PY - 2020
DO - 10.1587/transinf.2020EDL8016
JO - IEICE TRANSACTIONS on Information
SN - 1745-1361
VL - E103-D
IS - 8
JA - IEICE TRANSACTIONS on Information
Y1 - August 2020
AB - System logs record system states and significant events at various critical points to help debug performance issues and failures. Therefore, the rapid and accurate detection of the system log is crucial to the security and stability of the system. In this paper, proposed is a novel attention-based neural network model, which would learn log patterns from normal execution. Concretely, our model adopts a GRU module with attention mechanism to extract the comprehensive and intricate correlations and patterns embedded in a sequence of log entries. Experimental results demonstrate that our proposed approach is effective and achieve better performance than conventional methods.
ER -