IDEAS home Printed from https://ideas.repec.org/p/ant/wpaper/2018012.html
   My bibliography  Save this paper

Towards a scalable anomaly detection with pseudo-optimal hyperparameters

Author

Listed:
  • VANHOEYVELD, Jellis
  • MARTENS, David

Abstract

Anomaly detection (AD) involves the detection of instances that show a conduct that differs from a well de ned notion of normal behaviour. This apparently simple task still poses two open issues that are addressed in this work. (1) Each AD technique typically requires a number of hyperparameters to be specified. As labels are presumed unknown, supervised tuning approaches are ruled out. The community developed heuristics that are either too computationally involved and/or are tailored to a single type of AD method. Benchmarking studies comparing several AD methods therefore rely on an arbitrary parameter choice or adopt standard statistics (mean, min, max) to summarize the performances of a range of parameter settings. We consider the latter methodology as inappropriate and develop a heuristic to tune the hyperparameters of any AD algorithm. We choose the parameter combination that, on `average', achieves the highest performance on an external labelled data repository of AD tasks. The Wilcoxon-signed rank test reveals this setting to significantly outperform a random choice. (2) AD techniques can be designed with a focus on predictive performance and neglect scalability issues, limiting their applicability. We present the fixed-width anomaly detection (FWAD) method that conducts a preliminary data compression with fixed-width clustering and applies the original AD technique to the resulting cluster centres. The latter retain information of the original dataset in the form of local density estimates. Our results indicate that AD is significantly slower than FWAD if the former has a superlinear complexity in the number of instances N. Substantial bene ts are obtained for datasets of low-dimensionality with N large. Often, no significant AUC differences are observed between FWAD and AD, though we noted a drop in AUC for FWAD on a (small) majority of datasets. Apparently, the local density estimates hold enough information to allow for high-compression levels.

Suggested Citation

  • VANHOEYVELD, Jellis & MARTENS, David, 2018. "Towards a scalable anomaly detection with pseudo-optimal hyperparameters," Working Papers 2018012, University of Antwerp, Faculty of Business and Economics.
    • Handle: RePEc:ant:wpaper:2018012
    as

    Download full text from publisher

    File URL: https://repository.uantwerpen.be/docman/irua/450120/154651.pdf
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Markus Goldstein & Seiichi Uchida, 2016. "A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data," PLOS ONE, Public Library of Science, vol. 11(4), pages 1-31, April.
    Full references (including those not matched with items on IDEAS)

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Durgesh Samariya & Amit Thakkar, 2023. "A Comprehensive Survey of Anomaly Detection Algorithms," Annals of Data Science, Springer, vol. 10(3), pages 829-850, June.
    2. Vinicius Francisco Rofatto & Marcelo Tomio Matsuoka & Ivandro Klein & Maurício Roberto Veronez & Luiz Gonzaga da Silveira Junior, 2020. "On the effects of hard and soft equality constraints in the iterative outlier elimination procedure," PLOS ONE, Public Library of Science, vol. 15(8), pages 1-29, August.
    3. Yin, Sihua & Yang, Haidong & Xu, Kangkang & Zhu, Chengjiu & Zhang, Shaqing & Liu, Guosheng, 2022. "Dynamic real–time abnormal energy consumption detection and energy efficiency optimization analysis considering uncertainty," Applied Energy, Elsevier, vol. 307(C).
    4. Priyanga Dilini Talagala & Rob J Hyndman & Kate Smith-Miles, 2019. "Anomaly Detection in High Dimensional Data," Monash Econometrics and Business Statistics Working Papers 20/19, Monash University, Department of Econometrics and Business Statistics.
    5. Perwez, Usama & Yamaguchi, Yohei & Ma, Tao & Dai, Yanjun & Shimoda, Yoshiyuki, 2022. "Multi-scale GIS-synthetic hybrid approach for the development of commercial building stock energy model," Applied Energy, Elsevier, vol. 323(C).
    6. Sevvandi Kandanaarachchi & Mario A Munoz & Rob J Hyndman & Kate Smith-Miles, 2018. "On normalization and algorithm selection for unsupervised outlier detection," Monash Econometrics and Business Statistics Working Papers 16/18, Monash University, Department of Econometrics and Business Statistics.
    7. Taha Yehia & Ali Wahba & Sondos Mostafa & Omar Mahmoud, 2022. "Suitability of Different Machine Learning Outlier Detection Algorithms to Improve Shale Gas Production Data for Effective Decline Curve Analysis," Energies, MDPI, vol. 15(23), pages 1-25, November.
    8. Francisco Melo Pereira & Rute C. Sofia, 2022. "An Analysis of ML-Based Outlier Detection from Mobile Phone Trajectories," Future Internet, MDPI, vol. 15(1), pages 1-19, December.
    9. Irene Epifanio & Vicent Gimeno & Ximo Gual-Arnau & M. Victoria Ibáñez-Gual, 2020. "A New Geometric Metric in the Shape and Size Space of Curves in R n," Mathematics, MDPI, vol. 8(10), pages 1-19, October.
    10. Priyanga Dilini Talagala & Rob J Hyndman & Catherine Leigh & Kerrie Mengersen & Kate Smith-Miles, 2019. "A Feature-Based Framework for Detecting Technical Outliers in Water-Quality Data from In Situ Sensors," Monash Econometrics and Business Statistics Working Papers 1/19, Monash University, Department of Econometrics and Business Statistics.
    11. Piero Mazzarisi & Adele Ravagnani & Paola Deriu & Fabrizio Lillo & Francesca Medda & Antonio Russo, 2022. "A machine learning approach to support decision in insider trading detection," Papers 2212.05912, arXiv.org.
    12. Cian Ryan & Finbarr Murphy & Martin Mullins, 2019. "Semiautonomous Vehicle Risk Analysis: A Telematics‐Based Anomaly Detection Approach," Risk Analysis, John Wiley & Sons, vol. 39(5), pages 1125-1140, May.
    13. Elmira Asadi-Fard & Samereh Falahatkar & Mahdi Tanha Ziyarati & Xiaodong Zhang & Mariapia Faruolo, 2023. "Assessment of RXD Algorithm Capability for Gas Flaring Detection through OLI-SWIR Channels," Sustainability, MDPI, vol. 15(6), pages 1-20, March.
    14. Kenichiro Nagata & Toshikazu Tsuji & Kimitaka Suetsugu & Kayoko Muraoka & Hiroyuki Watanabe & Akiko Kanaya & Nobuaki Egashira & Ichiro Ieiri, 2021. "Detection of overdose and underdose prescriptions—An unsupervised machine learning approach," PLOS ONE, Public Library of Science, vol. 16(11), pages 1-14, November.
    15. Sevvandi Kandanaarachchi & Rob J Hyndman, 2019. "Dimension Reduction For Outlier Detection Using DOBIN," Monash Econometrics and Business Statistics Working Papers 17/19, Monash University, Department of Econometrics and Business Statistics.
    16. Timothy DeLise, 2023. "Deep Semi-Supervised Anomaly Detection for Finding Fraud in the Futures Market," Papers 2309.00088, arXiv.org.
    17. Ruhi Kiran Bajaj & Rebecca Mary Meiring & Fernando Beltran, 2023. "Co-Design, Development, and Evaluation of a Health Monitoring Tool Using Smartwatch Data: A Proof-of-Concept Study," Future Internet, MDPI, vol. 15(3), pages 1-15, March.
    18. Ke Lu & Xianwen Fang & Na Fang, 2022. "PN-BBN: A Petri Net-Based Bayesian Network for Anomalous Behavior Detection," Mathematics, MDPI, vol. 10(20), pages 1-24, October.
    19. Bruno Faria & Fernao Vistulo de Abreu, 2019. "Cellular frustration algorithms for anomaly detection applications," PLOS ONE, Public Library of Science, vol. 14(7), pages 1-31, July.
    20. Asoke K Nandi & Kuldeep Kaur Randhawa & Hong Siang Chua & Manjeevan Seera & Chee Peng Lim, 2022. "Credit card fraud detection using a hierarchical behavior-knowledge space model," PLOS ONE, Public Library of Science, vol. 17(1), pages 1-16, January.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:ant:wpaper:2018012. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Joeri Nys (email available below). General contact details of provider: https://edirc.repec.org/data/ftufsbe.html .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.