CoChat
Sign Up
Cited by 102Fault Detection and Control SystemsAnomaly Detection Techniques and ApplicationsMachine Fault Diagnosis Techniques

FLAGS: A methodology for adaptive anomaly detection and root cause analysis on sensor data streams by fusing expert knowledge with machine learning

Open access
Steenwinckel, Bram, De Paepe, Dieter, Vanden Hautte, Sander, Heyvaert, Pieter, Bentefrit, Mohamed, Moens, Pieter, Dimou, Anastasia, Van Den Bossche, Bruno +3 more·2021·Future Generation Computer Systems
DOI 10.1016/j.future.2020.10.015
Open this paper in CoChat — free accountOpen PDF

Abstract

Anomalies and faults can be detected, and their causes verified, using both data-driven and knowledge-driven techniques. Data-driven techniques can adapt their internal functioning based on the raw input data but fail to explain the manifestation of any detection. Knowledge-driven techniques inherently deliver the cause of the faults that were detected but require too much human effort to set up. In this paper, we introduce FLAGS, the Fused-AI interpretabLe Anomaly Generation System, and combine both techniques in one methodology to overcome their limitations and optimize them based on limited user feedback. Semantic knowledge is incorporated in a machine learning technique to enhance expressivity. At the same time, feedback about the faults and anomalies that occurred is provided as input to increase adaptiveness using semantic rule mining methods. This new methodology is evaluated on a predictive maintenance case for trains. We show that our method reduces their downtime and provides more insight into frequently occurring problems.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1016/j.future.2020.10.015

Cites

66 of 66 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Having explicit features makes INK embeddings interpretable and usable for root cause analysis.

    “Furthermore, having explicit features makes the INK embeddings interpretable and usable for root cause analysis [60].”

    From Toward Context-Aware Anomaly Detection for AIOps in Microservices Using Dynamic Knowledge Graphs (2026) · cites this paper as [60]

    Section 3.1 says that mapping anomalies and causes into a knowledge graph adds interpretability, and Section 7 discusses using FLAGS to investigate anomalies. Searches for “INK” and “explicit features embeddings” found no passage about INK embeddings or whether their features enable root cause analysis. (LLM, conf=0.75)

    Experimental — accuracy not yet measured

  • The use of INK as a node embedding technique inherently supports explainability and root cause analysis.

    “While quantitative evaluation of explainability and RCA has not been the focus of this work, the use of INK as a node embedding technique inherently supports both [60].”

    From Toward Context-Aware Anomaly Detection for AIOps in Microservices Using Dynamic Knowledge Graphs (2026) · cites this paper as [60]

    The Discussion (section 7) attributes interpretability to FLAGS’ semantic output and fault detector, while section 3.1 describes a knowledge graph that adds interpretability to anomalies and their causes. Searches for INK and node embeddings found no passage saying that INK, as an embedding technique, supports explainability or root cause analysis. (LLM, conf=0.76)

    Experimental — accuracy not yet measured

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 2

  1. S.K. Bose, B. Kar, M. Roy, P.K. Gopalakrishnan, A. Basu, ADEPOS: anomaly detection based power saving for predictive maintenance using edge computing, in: Proceedings of the 24th Asia and South Pacific Design Automation Conference, 2019, pp. 597-602.
  2. P. Umiliacchi, R. Shingler, G. Langer, U. Henning, A new approach to opti- misation through intelligent integration of railway systems: the InteGRail project, in: Proceedings of the 7th WCRR, World Conference on Railway Research, 2006.

Cited by

Sorted by citations

Related

Authors
Steenwinckel, B., De Paepe, D., Vanden Hautte, S., Heyvaert, P., Bentefrit, M., Moens, P., Dimou, A., Van Den Bossche, B., De Turck, F., Van Hoecke, S., & Ongenae, F.
Affiliations
Ghent University (BE)
Corresponding author
Steenwinckel, Bram
ORCID
Venue
Future Generation Computer Systems
Publisher
Elsevier BV
Published
2021-03
Volume
116
Pages
30-48
ISSN
0167-739X (ISSN-L)
DOI
10.1016/j.future.2020.10.015
OpenAlex
W3094146027
Cited by
102
References
66
Citations by year
2026: 13 · 2025: 17 · 2024: 22 · 2023: 17 · 2022: 18 · 2021: 15
FWCI
6.18
Type
journal article
Language
en
Open access
Licence
Locations
  • Future Generation Computer Systems (publishedVersion, cc-by)
  • Ghent University Academic Bibliography (Ghent University) (submittedVersion, cc-by)
Funders
Agentschap Innoveren en Ondernemen
Topics
  • Physical Sciences › Engineering › Control and Systems Engineering › Fault Detection and Control Systems
  • Physical Sciences › Computer Science › Artificial Intelligence › Anomaly Detection Techniques and Applications
  • Physical Sciences › Engineering › Control and Systems Engineering › Machine Fault Diagnosis Techniques
Keywords
anomaly detection, root cause analysis, sensor data streams, model interpretability, user feedback, adaptive fault detection, predictive maintenance, train maintenance

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review