Academic Bibliography

 Search 200 years of publications by Ghent University researchers.
Advanced search
2 files | 2.58 MB
Add to list
  1. Search results
  2. Adaptive anomaly detection and root c...

Adaptive anomaly detection and root cause analysis by fusing semantics and machine learning

Bram Steenwinckel (UGent)
(2018) SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS. In Lecture notes in computer science 11155. p.272-282
Author
Organization
Abstract
Anomaly detection (AD) systems are either manually built by experts setting thresholds on data or constructed automatically by learning from the available data through machine learning (ML). The first requires profound prior knowledge and are non-adaptive to changing environments but can perform root cause analysis (RCA) to give an understanding of the detected anomaly. The second has a huge need for data, is unable to perform RCA and is often only trained once and deployed in various contexts, leading to a lot of false positives. Fusing the prior knowledge with ML techniques could resolve the generation of these alarms and should define the causes. The primary challenges to create such a detection system are: (1) Augmenting the current ML techniques with prior knowledge to enhance the detection rate. (2) Incorporate knowledge to interpret the cause of a detected anomaly automatically. (3) Reduce of human-involvement by automating the design of detection patterns.
Keywords
Anomaly detection, Root cause analysis, Machine learning, Expert knowledge, Semantic web, Knowledge graphs, CONTEXT

Downloads

  • Download
    (...).pdf
    • full text (Published version)
    • |
    • UGent only
    • |
    • PDF
    • |
    • 908.23 KB
  • Download
    paper adaptive.pdf
    • full text (Accepted manuscript)
    • |
    • open access
    • |
    • PDF
    • |
    • 1.67 MB

Citation

Please use this url to cite or link to this publication:

MLA
Steenwinckel, Bram. “Adaptive Anomaly Detection and Root Cause Analysis by Fusing Semantics and Machine Learning.” SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS, edited by Aldo Gangemi et al., vol. 11155, Springer, 2018, pp. 272–82, doi:10.1007/978-3-319-98192-5_46.
APA
Steenwinckel, B. (2018). Adaptive anomaly detection and root cause analysis by fusing semantics and machine learning. In A. Gangemi, A. L. Gentile, A. G. Nuzzolese, S. Rudolph, M. Maleshkova, H. Paulheim, … M. Alam (Eds.), SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS (Vol. 11155, pp. 272–282). https://doi.org/10.1007/978-3-319-98192-5_46
Chicago author-date
Steenwinckel, Bram. 2018. “Adaptive Anomaly Detection and Root Cause Analysis by Fusing Semantics and Machine Learning.” In SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS, edited by Aldo Gangemi, Anna Lisa Gentile, Andrea Giovanni Nuzzolese, Sebastian Rudolph, Maria Maleshkova, Heiko Paulheim, Jeff Z. Pan, and Mehwish Alam, 11155:272–82. Springer. https://doi.org/10.1007/978-3-319-98192-5_46.
Chicago author-date (all authors)
Steenwinckel, Bram. 2018. “Adaptive Anomaly Detection and Root Cause Analysis by Fusing Semantics and Machine Learning.” In SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS, ed by. Aldo Gangemi, Anna Lisa Gentile, Andrea Giovanni Nuzzolese, Sebastian Rudolph, Maria Maleshkova, Heiko Paulheim, Jeff Z. Pan, and Mehwish Alam, 11155:272–282. Springer. doi:10.1007/978-3-319-98192-5_46.
Vancouver
1.
Steenwinckel B. Adaptive anomaly detection and root cause analysis by fusing semantics and machine learning. In: Gangemi A, Gentile AL, Nuzzolese AG, Rudolph S, Maleshkova M, Paulheim H, et al., editors. SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS. Springer; 2018. p. 272–82.
IEEE
[1]
B. Steenwinckel, “Adaptive anomaly detection and root cause analysis by fusing semantics and machine learning,” in SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS, Heraklion, GREECE, 2018, vol. 11155, pp. 272–282.
@inproceedings{8667066,
  abstract     = {{Anomaly detection (AD) systems are either manually built by experts setting thresholds on data or constructed automatically by learning from the available data through machine learning (ML). The first requires profound prior knowledge and are non-adaptive to changing environments but can perform root cause analysis (RCA) to give an understanding of the detected anomaly. The second has a huge need for data, is unable to perform RCA and is often only trained once and deployed in various contexts, leading to a lot of false positives. Fusing the prior knowledge with ML techniques could resolve the generation of these alarms and should define the causes. The primary challenges to create such a detection system are: (1) Augmenting the current ML techniques with prior knowledge to enhance the detection rate. (2) Incorporate knowledge to interpret the cause of a detected anomaly automatically. (3) Reduce of human-involvement by automating the design of detection patterns.}},
  author       = {{Steenwinckel, Bram}},
  booktitle    = {{SEMANTIC WEB: ESWC 2018 SATELLITE EVENTS}},
  editor       = {{Gangemi, Aldo and Gentile, Anna Lisa and Nuzzolese, Andrea Giovanni and Rudolph, Sebastian and Maleshkova, Maria and Paulheim, Heiko and Pan, Jeff Z. and Alam, Mehwish}},
  isbn         = {{9783319981918}},
  issn         = {{0302-9743}},
  keywords     = {{Anomaly detection,Root cause analysis,Machine learning,Expert knowledge,Semantic web,Knowledge graphs,CONTEXT}},
  language     = {{eng}},
  location     = {{Heraklion, GREECE}},
  pages        = {{272--282}},
  publisher    = {{Springer}},
  title        = {{Adaptive anomaly detection and root cause analysis by fusing semantics and machine learning}},
  url          = {{http://doi.org/10.1007/978-3-319-98192-5_46}},
  volume       = {{11155}},
  year         = {{2018}},
}
Altmetric
View in Altmetric
Web of Science
Times cited: