Robust and Transferable Anomaly Detection in Log Data using Pre-Trained Language Models

被引:9
|
作者
Ott, Harold [1 ]
Bogatinovski, Jasmin [1 ]
Acker, Alexander [1 ]
Nedelkoski, Sasho [1 ]
Kao, Odej [1 ]
机构
[1] TU Berlin, Distributed & Operating Syst, Berlin, Germany
关键词
anomaly detection; log analysis; deep learning; language models; transfer learning;
D O I
10.1109/CloudIntelligence52565.2021.00013
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Anomalies or failures in large computer systems, such as the cloud, have an impact on a large number of users that communicate, compute, and store information. Therefore, timely and accurate anomaly detection is necessary for reliability, security, safe operation, and mitigation of losses in these increasingly important systems. Recently, the evolution of the software industry opens up several problems that need to be tackled including (1) addressing the software evolution due software upgrades, and (2) solving the cold-start problem, where data from the system of interest is not available. In this paper, we propose a framework for anomaly detection in log data, as a major troubleshooting source of system information. To that end, we utilize pre-trained general-purpose language models to preserve the semantics of log messages and map them into log vector embeddings. The key idea is that these representations for the logs are robust and less invariant to changes in the logs, and therefore, result in a better generalization of the anomaly detection models. We perform several experiments on a cloud dataset evaluating different language models for obtaining numerical log representations such as BERT, GPT-2, and XL. The robustness is evaluated by gradually altering log messages, to simulate a change in semantics. Our results show that the proposed approach achieves high performance and robustness, which opens up possibilities for future research in this direction.
引用
收藏
页码:19 / 24
页数:6
相关论文
共 50 条
  • [1] BERT-Log: Anomaly Detection for System Logs Based on Pre-trained Language Model
    Chen, Song
    Liao, Hai
    APPLIED ARTIFICIAL INTELLIGENCE, 2022, 36 (01)
  • [2] Robust Lottery Tickets for Pre-trained Language Models
    Zheng, Rui
    Bao, Rong
    Zhou, Yuhao
    Liang, Di
    Wane, Sirui
    Wu, Wei
    Gui, Tao
    Zhang, Qi
    Huang, Xuanjing
    PROCEEDINGS OF THE 60TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022), VOL 1: (LONG PAPERS), 2022, : 2211 - 2224
  • [3] Sprelog: Log-Based Anomaly Detection with Self-matching Networks and Pre-trained Models
    Yang, Haitian
    Zhao, Xuan
    Sun, Degang
    Wang, Yan
    Huang, Weiqing
    SERVICE-ORIENTED COMPUTING (ICSOC 2021), 2021, 13121 : 736 - 743
  • [4] Emotional Paraphrasing Using Pre-trained Language Models
    Casas, Jacky
    Torche, Samuel
    Daher, Karl
    Mugellini, Elena
    Abou Khaled, Omar
    2021 9TH INTERNATIONAL CONFERENCE ON AFFECTIVE COMPUTING AND INTELLIGENT INTERACTION WORKSHOPS AND DEMOS (ACIIW), 2021,
  • [5] Pre-Trained Language Models and Their Applications
    Wang, Haifeng
    Li, Jiwei
    Wu, Hua
    Hovy, Eduard
    Sun, Yu
    ENGINEERING, 2023, 25 : 51 - 65
  • [6] Adapting Pre-trained Language Models to Rumor Detection on Twitter
    Slimi, Hamda
    Bounhas, Ibrahim
    Slimani, Yahya
    JOURNAL OF UNIVERSAL COMPUTER SCIENCE, 2021, 27 (10) : 1128 - 1148
  • [7] A Data Cartography based MixUp for Pre-trained Language Models
    Park, Seo Yeon
    Caragea, Cornelia
    NAACL 2022: THE 2022 CONFERENCE OF THE NORTH AMERICAN CHAPTER OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS: HUMAN LANGUAGE TECHNOLOGIES, 2022, : 4244 - 4250
  • [8] Pre-trained Language Models with Limited Data for Intent Classification
    Kasthuriarachchy, Buddhika
    Chetty, Madhu
    Karmakar, Gour
    Walls, Darren
    2020 INTERNATIONAL JOINT CONFERENCE ON NEURAL NETWORKS (IJCNN), 2020,
  • [9] μBERT: Mutation Testing using Pre-Trained Language Models
    Degiovanni, Renzo
    Papadakis, Mike
    2022 IEEE 15TH INTERNATIONAL CONFERENCE ON SOFTWARE TESTING, VERIFICATION AND VALIDATION WORKSHOPS (ICSTW 2022), 2022, : 160 - 169
  • [10] Devulgarization of Polish Texts Using Pre-trained Language Models
    Klamra, Cezary
    Wojdyga, Grzegorz
    Zurowski, Sebastian
    Rosalska, Paulina
    Kozlowska, Matylda
    Ogrodniczuk, Maciej
    COMPUTATIONAL SCIENCE, ICCS 2022, PT II, 2022, : 49 - 55