Is your anomaly detector ready for change? adapting aiops solutions to the real world

Lorena Poenaru-Olaru*, Natalia Karpova, Luis Cruz, Jan S. Rellermeyer, Arie Van Deursen

*Corresponding author for this work

Research output: Chapter in Book/Conference proceedings/Edited volumeConference contributionScientificpeer-review

Abstract

Anomaly detection techniques are essential in automating the monitoring of IT systems and operations. These techniques imply that machine learning algorithms are trained on operational data corresponding to a specific period of time and that they are continuously evaluated on newly emerging data. Operational data is constantly changing over time, which affects the performance of deployed anomaly detection models. Therefore, continuous model maintenance is required to preserve the performance of anomaly detectors over time. In this work, we analyze two different anomaly detection model maintenance techniques in terms of the model update frequency, namely blind model retraining and informed model retraining. We further investigate the effects of updating the model by retraining it on all the available data (full-history approach) and only the newest data (sliding window approach). Moreover, we investigate whether a data change monitoring tool is capable of determining when the anomaly detection model needs to be updated through retraining.

Original languageEnglish
Title of host publicationProceedings - 2024 IEEE/ACM 3rd International Conference on AI Engineering - Software Engineering for AI, CAIN 2024
PublisherAssociation for Computing Machinery (ACM)
Pages222-233
Number of pages12
ISBN (Electronic)9798400705915
DOIs
Publication statusPublished - 2024
Event3rd International Conference on AI Engineering, CAIN 2024, co-located with the 46th International Conference on Software Engineering, ICSE 2024 - Lisbon, Portugal
Duration: 14 Apr 202415 Apr 2024

Publication series

NameProceedings - 2024 IEEE/ACM 3rd International Conference on AI Engineering - Software Engineering for AI, CAIN 2024

Conference

Conference3rd International Conference on AI Engineering, CAIN 2024, co-located with the 46th International Conference on Software Engineering, ICSE 2024
Country/TerritoryPortugal
CityLisbon
Period14/04/2415/04/24

Keywords

  • AIOps
  • anomaly detection
  • concept drift detection
  • model maintenance
  • model monitoring

Fingerprint

Dive into the research topics of 'Is your anomaly detector ready for change? adapting aiops solutions to the real world'. Together they form a unique fingerprint.

Cite this