The DOI matches its Crossref record; 228 of the 228 references it cites were found in a registry or an index; 45 of the 131 sources its claims cite were checked for support.
A Survey on Explainable Anomaly Detection
Open accessAbstract
This survey provides a comprehensive and structured survey of state-of-the-art explainable anomaly detection techniques.
Overview
CoChat Summary
TL;DR
This paper is a survey of explainable anomaly detection (XAD) that reviews more than 150 articles and proposes a six-criteria taxonomy to organize the field. It argues that XAD is needed because anomaly detection is widely used in high-stakes settings, but explanations must be faithful, comprehensible, scalable, and aligned with the actual anomaly definition.
Key findings
- The survey reviews more than 150 articles on XAD and organizes them into four high-level categories and twelve fine-grained categories.
- It proposes six main taxonomy criteria for XAD methods: explanation type, model-agnostic vs. model-specific, data type, local vs. global explanation, techniques used, and applicable data type.
- The authors report that pre-model techniques are generally model-agnostic and provide global explanations, while in-model techniques are model-specific and often provide feature-based explanations.
- They note that post-model techniques are often local and can suffer from poor explanation fidelity, especially when the explanation definition differs from the detection definition.
Methods
- The paper is a comparative, iterative literature survey with three cycles of collection and screening.
- It searches Google Scholar, IEEE Xplore, ACM Digital Library, DBLP, and Web of Science, covering publications from January 1998 to February 2022.
- The survey classifies methods using a taxonomy built from six criteria and then reviews the literature in four sections: pre-model, in-model, shallow post-model, and deep post-model techniques.
Limitations
- The paper states there is no consensus on what a good XAD technique should be, especially for evaluating explanation quality.
- It states that evaluation of XAD remains unresolved, and that methods developed for classification or clustering are non-trivial to extend to anomaly detection.
- Evident from the text: because the survey is based on literature up to February 2022 and explicitly says it does not claim to have covered all relevant publications, it may not include later work or every relevant method.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
20 Valid · 48 Not valid · 74 Not checked · 116 Own results and interpretations
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“Specifically, Pang et al. [156] and Pang et al. [158] propose two filter-based unsupervised selection methods, namely, CBRW_FS and CBRW, which select a subset of features independently from subsequent anomaly detection methods.”
§ A Survey on Explainable Anomaly Detection > 4.1 Feature Selection for Anomaly Detection
What the sources say
- Not matchedcites [156] 156. Guansong Pang and Charu Aggarwal. 2021. Toward explainable deep anomaly detection. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 4056-4057.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Disputedcites [158] 158. Guansong Pang, Longbing Cao, Ling Chen, Defu Lian, and Huan Liu. 2018. Sparse modeling-based sequential en- semble learning for effective outlier detection in high-dimensional numeric data. In Proceedings of the 32nd AAAI Conference on Artificial Intelligence. sourceExperimental — accuracy not yet measured
The Abstract (p. 1) says this paper introduces SEMSE and CINFO, which iteratively connect feature selection with outlier scoring rather than selecting features independently. The Introduction (p. 1) describes independent feature selection as a limitation of earlier methods. Our copy does not identify CBRW_FS or CBRW as methods proposed in this paper. (LLM, conf=0.87)
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
At least one cited source doesn’t support this claim or points to a different paper, and none confirmed it — see the verdicts below.
In the paper
“As shown in Table 1, all pre-model XAD techniques are model-agnostic except for Reference [173].”
§ A Survey on Explainable Anomaly Detection > 4.3 Summary
What the sources say
- Disputedcites [173] 173. Sridhar Ramaswamy, Rajeev Rastogi, and Kyuseok Shim. 2000. Efficient algorithms for mining outliers from large data sets. In Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data. 427-438. sourceExperimental — accuracy not yet measured
The Introduction describes this paper’s method as distance-based and contrasts it with approaches requiring a postulated statistical model. Related Work says its outlier definition is independent of how clusters are identified. That runs contrary to treating this paper as the exception to model-agnostic pre-model techniques; the paper does not assess the other techniques in the citing document’s Table 1. (LLM, conf=0.84)
Checked by a reasoning model — may be wrong.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Disputed
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“These two methods work only on categorical data through modelling the feature-value couplings.”
§ A Survey on Explainable Anomaly Detection > 4.1 Feature Selection for Anomaly Detection
What the sources say
- Not matchedcites [156] 156. Guansong Pang and Charu Aggarwal. 2021. Toward explainable deep anomaly detection. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining. 4056-4057.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Partly supportedcites [158] 158. Guansong Pang, Longbing Cao, Ling Chen, Defu Lian, and Huan Liu. 2018. Sparse modeling-based sequential en- semble learning for effective outlier detection in high-dimensional numeric data. In Proceedings of the 32nd AAAI Conference on Artificial Intelligence. sourceExperimental — accuracy not yet measured
The High-dimensional Outlier Detection Methods section (p. 2) describes a group of methods for categorical data and says attempts to adapt them to numeric data by discretization performed poorly. The References (p. 8) include works on feature-value couplings, but the paper does not identify CBRW_FS or CBRW or establish that both work only through those couplings. (LLM, conf=0.77)
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
The sources with a verdict support this claim; the rest weren’t checked yet, have no full text we could read, or their verdicts aren’t rated yet.
In the paper
“Compared to the abundance of taxonomies of anomaly detection methods, including but not limited to these surveys throughout the years [3,36,37,78,86,161], the categorization of anomaly explanation methods involving XAD techniques has received relatively little attention so far [162,185,189,216,225].”
§ A Survey on Explainable Anomaly Detection > 3.1 Related Work
What the sources say
- Not matchedcites [162] 162. Guansong Pang, Chunhua Shen, Longbing Cao, and Anton Van Den Hengel. 2021. Deep learning for anomaly detec- tion: A review. ACM Computing Surveys (CSUR) 54, 2 (2021), 1-38.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
- Supportedcites [185] 185. Durgesh Samariya, Sunil Aryal, Kai Ming Ting, and Jiangang Ma. 2020. A new effective and efficient measure for out- lying aspect mining. In Proceedings of the International Conference on Web Information Systems Engineering. Springer, 463-474. source
The Introduction (p. 2) says that, to the authors’ knowledge, no survey of outlying aspect mining (OAM) had been conducted and presents this paper’s categorization of OAM techniques. The Introduction (p. 1) identifies OAM as a form of outlier explanation. Together, these passages support the cited point that categorization of anomaly-explanation methods had received relatively little attention. (LLM, conf=0.88)
- Not checked — no full text of the sourcecites [189] 189. Bernhard Schölkopf, Robert C. Williamson, Alex Smola, John Shawe-Taylor, and John Platt. 1999. Support vector method for novelty detection. In Proceedings of the 12th International Conference on Neural Information Processing Systems (Denver, CO) (NIPS'99, Vol. 12). MIT Press, Cambridge, MA, USA, 582-588. source
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Not matched
The cited sources support only part of this claim — read each source’s verdict below.
In the paper
“Specifically, each of the above surveys contains no more than 40 relevant works in the field.”
§ A Survey on Explainable Anomaly Detection > 3.1 Related Work
What the sources say
- Could not verifycites [216] 216. Shashanka Venkataramanan, Kuan-Chuan Peng, Rajat Vikram Singh, and Abhijit Mahalanobis. 2020. Attention guided anomaly localization in images. In Proceedings of the European Conference on Computer Vision. Springer, 485-503. sourceExperimental — accuracy not yet measured
The Related Works section (p. 4) compares anomaly-localization methods, but does not count works covered by XAD surveys. Searches for surveys, XAD, and the 40-work limit found no passage in our copy addressing the claim. (LLM, conf=0.76)
- Supportedcites [185] 185. Durgesh Samariya, Sunil Aryal, Kai Ming Ting, and Jiangang Ma. 2020. A new effective and efficient measure for out- lying aspect mining. In Proceedings of the International Conference on Web Information Systems Engineering. Springer, 463-474. source
For this cited survey, the References (pp. 11–12) are numbered through 16, and Open Challenges (p. 10) cites a further work as [17]. Thus the paper contains fewer than 40 referenced works, consistent with the claim as applied to this survey. It does not establish the counts for the other surveys mentioned in the citing sentence. (LLM, conf=0.85)
- Not matchedcites [162] 162. Guansong Pang, Chunhua Shen, Longbing Cao, and Anton Van Den Hengel. 2021. Deep learning for anomaly detec- tion: A review. ACM Computing Surveys (CSUR) 54, 2 (2021), 1-38.
Couldn’t be matched to a paper in Crossref, so it can’t be checked.
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
- Could not verify
118 more claims · 16 with their cited sources checked
Create a free account to see all 258 claims and every source verdict.
- Probably not a substantial claim. 108
- It may state two separate claims as one. 34
- The quote only partly states this claim. 19
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
Each contradiction between two passages listed here was confirmed by an independent check.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ A Survey on Explainable Anomaly Detection > 7.3 Explaining Convolutional Neural Networks
“Szymanowicz et al. [207] introduce a model to detect and explain anomalies in videos.”
§ A Survey on Explainable Anomaly Detection > References
“207. Mukund Sundararajan, Ankur Taly, and Qiqi Yan. 2017. Axiomatic attribution for deep networks. In Proceedings of the International Conference on Machine Learning. PMLR, 3319-3328.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ A Survey on Explainable Anomaly Detection > 7.2 Explaining Recurrent Neural Networks
“Patil et al. [166] use LSTM to detect anomalies in such systems.”
§ A Survey on Explainable Anomaly Detection > References
“166. Animesh Patcha and Jung-Min Park. 2007. An overview of anomaly detection techniques: Existing solutions and latest technological trends. Computer Networks 51, 12 (2007), 3448-3470.”
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
§ A Survey on Explainable Anomaly Detection > 4.2 Feature Representation for Anomaly Detection
“Ramirez et al. [173] introduce an interpretable anomaly detection and classification framework to analyze human gait.”
§ A Survey on Explainable Anomaly Detection > References
“173. Sridhar Ramaswamy, Rajeev Rastogi, and Kyuseok Shim. 2000. Efficient algorithms for mining outliers from large data sets. In Proceedings of the 2000 ACM SIGMOD International Conference on Management of Data. 427-438.”
Citations
Cites
U-Net: Convolutional Networks for Biomedical Image Segmentation
Ronneberger, O., Fischer, P., & Brox, T. (2015) · Lecture notes in computer science · doi:10.1007/978-3-319-24574-4_28 · Cited by 99,404 · cited 2×
“They first design an encoder-decoder architecture based on U-Net [175] to detect anomalies, thereby generating saliency maps by computing per-pixel differences between actual and predicted frames.”
MethodXGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · cited 1×
“Second, they utilize two types of surrogate models to explain anomalies, including black-box anomaly detection models with a post-hoc local explanation (XGBoost [44] and LightGBM [98] with LIME or SHAP), and transparent anomaly detection models (ElasticNet [233] and EBM [153]).”
Method(2008) · Journal of Machine Learning Research · Cited by 33,051
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.74 · Cited by 24,513
Regularization and Variable Selection Via the Elastic Net
Zou, H., & Hastie, T. (2005) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2005.00503.x · Cited by 20,957
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Visualizing and understanding convolutional networks
Zeiler, M. D., & Fergus, R. (2014) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-10590-1_53 · Cited by 15,839 · cited 1×
“Moreover, to interpret an anomaly, they combine Shapley values and Occlusion maps [228] to investigate how input features impact the prediction.”
MethodChandola, V., Banerjee, A., & Kumar, V. (2009) · ACM Computing Surveys · doi:10.1145/1541880.1541882 · Cited by 11,687 · cited 10×
“Since the seminal work in [105], anomaly detection has been well studied and there exists a plethora of comprehensive surveys and reviews on it, including but not limited to References [1, 5, 25, 36, 37, 134, 135, 162, 166, 232].”
BackgroundInfluentialRobust Locally Weighted Regression and Smoothing Scatterplots
Cleveland, W. S. (1979) · Journal of the American Statistical Association · doi:10.1080/01621459.1979.10481038 · Cited by 11,594 · cited 1×
“For example, for each data instance, Reference [40] applies LOESS regression [48] by taking each feature in turn as the target variable and the remaining features as predictors based on its neighbours.”
MethodArrieta, A. B., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., García, S., Gil-López, S., Molina, D., Benjamins, R., et al. (2019) · arXiv · doi:10.48550/arxiv.1910.10045 · Cited by 11,321
Rudin, C. (2019) · Nature Machine Intelligence · doi:10.1038/s42256-019-0048-x · Cited by 10,561 · cited 1×
“For this reason, some researchers [181] appeal not to use opaque models and then explain them in a post-hoc manner for high-stakes decisions.”
BackgroundBarredo Arrieta, A., Díaz-Rodríguez, N., Del Ser, J., Bennetot, A., Tabik, S., Barbado, A., Garcia, S., Gil-Lopez, S., Molina, D., Benjamins, R., et al. (2020) · Information Fusion · doi:10.1016/j.inffus.2019.12.012 · Cited by 10,423 · cited 22×
“After carefully scrutinizing existing surveys on XAI [13, 20, 30, 34, 62, 71, 120], we found that some criteria are often used to categorize existing XAI techniques.”
BackgroundMethodInfluential“Why Should I Trust You?”: Explaining the Predictions of Any Classifier
Ribeiro, M., Singh, S., & Guestrin, C. (2016) · Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Demonstrations · doi:10.18653/v1/n16-3020 · Cited by 6,682
Liu, F. T., Ting, K. M., & Zhou, Z.-H. (2008) · 2008 Eighth IEEE International Conference on Data Mining · doi:10.1109/icdm.2008.17 · Cited by 6,601 · cited 5×
“[195] apply Isolation Forest [122] to detect cyber attacks and leverages SFEs to generate explanations.”
BackgroundMethodInfluentialLipton, Z. C. (2018) · Queue · doi:10.1145/3236386.3241340 · Cited by 5,427
Explanation in artificial intelligence: Insights from the social sciences
Miller, T. (2019) · Artificial Intelligence · doi:10.1016/j.artint.2018.07.007 · Cited by 5,302 · cited 2×
“Miller [142] defined XAI as a human-agent interaction problem at the intersection of Artificial Intelligence, HumanComputer Interaction (HCI), and the Social Sciences (including Philosophy, Cognitive Science, and Social Psychology).”
BackgroundLOF: identifying density-based local outliers
Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000) · ACM SIGMOD Record · doi:10.1145/335191.335388 · Cited by 5,207
LOF: identifying density-based local outliers
Breunig, M. M., Kriegel, H.-P., Ng, R. T., & Sander, J. (2000) · Proceedings of the 2000 ACM SIGMOD international conference on Management of data · doi:10.1145/342009.335388 · Cited by 4,149 · cited 10×
“For example, KNN [170] defines objects with ‘far’ 𝑘-nearest neighbours as anomalies, LOF [26] treats objects with a low local density as anomalies, and Isolation Forest [121] considers ‘easily isolated’ objects as anomalies.”
BackgroundMethodInfluentialA Survey of Outlier Detection Methodologies
Hodge, V., & Austin, J. (2004) · Artificial Intelligence Review · doi:10.1007/s10462-004-4304-y · Cited by 3,596
Striving for Simplicity: The All Convolutional Net
Springenberg, J. T., Dosovitskiy, A., Brox, T., & Riedmiller, M. (2014) · arXiv · doi:10.48550/arxiv.1412.6806 · Cited by 3,447
Towards A Rigorous Science of Interpretable Machine Learning
Doshi-Velez, F., & Kim, B. (2017) · arXiv · doi:10.48550/arxiv.1702.08608 · Cited by 3,121
The EU General Data Protection Regulation (GDPR)
Voigt, P., & von dem Bussche, A. (2017) · doi:10.1007/978-3-319-57959-7 · Cited by 3,001
Hawkins, D. M. (1980) · doi:10.1007/978-94-015-3994-4 · Cited by 2,982
Explainable AI: a review of machine learning interpretability methods
Linardatos, P., Papastefanopoulos, V., & Kotsiantis, S. (2020) · Entropy · doi:10.3390/e23010018 · Cited by 2,922 · cited 20×
“Meanwhile, a refined taxonomy, distilled from existing surveys on XAI techniques, is presented below and used to categorize XAD methods.”
MethodInfluentialProceedings of the 1999 American Control Conference
(1999) · Proceedings of the 1999 American Control Conference (Cat. No. 99CH36251) · doi:10.1109/acc.1999.782304 · Cited by 2,898
Deep Learning for Anomaly Detection: A Review
Pang, G., Shen, C., Cao, L., & Hengel, A. V. D. (2022) · ACM Computing Surveys · doi:10.1145/3439950 · Cited by 2,837
Methods for interpreting and understanding deep neural networks
Montavon, G., Samek, W., & Müller, K.-R. (2018) · Digital Signal Processing · doi:10.1016/j.dsp.2017.10.011 · Cited by 2,665
Explaining Explanations: An Overview of Interpretability of Machine Learning
Gilpin, L. H., Bau, D., Yuan, B. Z., Bajwa, A., Specter, M., & Kagal, L. (2018) · 2018 IEEE 5th International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa.2018.00018 · Cited by 2,507 · cited 21×
“After carefully scrutinizing existing surveys on XAI [13, 20, 30, 34, 62, 71, 120], we found that some criteria are often used to categorize existing XAI techniques.”
MethodInfluentialDefinitions, methods, and applications in interpretable machine learning
Murdoch, W. J., Singh, C., Kumbier, K., Abbasi-Asl, R., & Yu, B. (2019) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.1900654116 · Cited by 2,167 · cited 2×
“For instance, Reference [147] points out that combining exploratory data analysis tools with domain knowledge is helpful for extracting meaningful features, thereby improving the interpretbility of subsequent anomaly detection.”
BackgroundAnchors: High-Precision Model-Agnostic Explanations
Ribeiro, M. T., Singh, S., & Guestrin, C. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11491 · Cited by 2,162 · cited 4×
“Examples include Anchors [173].”
MethodInfluentialEfficient algorithms for mining outliers from large data sets
Ramaswamy, S., Rastogi, R., & Shim, K. (2000) · ACM SIGMOD Record · doi:10.1145/335191.335437 · Cited by 1,860
Machine Learning Interpretability: A Survey on Methods and Metrics
Carvalho, D. V., Pereira, E. M., & Cardoso, J. S. (2019) · Electronics · doi:10.3390/electronics8080832 · Cited by 1,842 · cited 20×
“Meanwhile, a refined taxonomy, distilled from existing surveys on XAI techniques, is presented below and used to categorize XAD methods.”
MethodInfluentialDeepLog: Anomaly Detection and Diagnosis from System Logs through Deep Learning
Du, M., Li, F., Zheng, G., & Srikumar, V. (2017) · Proceedings of the 2017 ACM SIGSAC Conference on Computer and Communications Security · doi:10.1145/3133956.3134015 · Cited by 1,835 · cited 2×
“For instance, Reference [205] utilises Decision Trees [41] and DeepLog [62] to detect anomalies in system logs, and then explain the results using the Shapley value approach.”
MethodA survey of network anomaly detection techniques
Ahmed, M., Naser Mahmood, A., & Hu, J. (2016) · Journal of Network and Computer Applications · doi:10.1016/j.jnca.2015.11.016 · Cited by 1,750
Deep Learning for Anomaly Detection: A Review
Pang, G., Shen, C., Cao, L., et al. (2020) · Singapore Management University Institutional Knowledge (InK) (Singapore Management University) · doi:10.48550/arxiv.2007.02500 · Cited by 1,718
An overview of anomaly detection techniques: Existing solutions and latest technological trends
Patcha, A., & Park, J.-M. (2007) · Computer Networks · doi:10.1016/j.comnet.2007.02.001 · Cited by 1,561 · cited 9×
“Since the seminal work in [105], anomaly detection has been well studied and there exists a plethora of comprehensive surveys and reviews on it, including but not limited to References [1, 5, 25, 36, 37, 134, 135, 162, 166, 232].”
BackgroundInfluentialSmoothGrad: removing noise by adding noise
Smilkov, D., Thorat, N., Kim, B., Viégas, F., & Wattenberg, M. (2017) · arXiv · doi:10.48550/arxiv.1706.03825 · Cited by 1,467 · cited 2×
“Next, they employ the GradientExplainer tool [127], which combines SHAP, Integrated Gradients [207], and SmoothGrad [202], to analyze which input features are associated with the anomaly.”
MethodNovelty detection: a review—part 1: statistical approaches
Markou, M., & Singh, S. (2003) · Signal Processing · doi:10.1016/j.sigpro.2003.07.018 · Cited by 1,432 · cited 9×
“Since the seminal work in [105], anomaly detection has been well studied and there exists a plethora of comprehensive surveys and reviews on it, including but not limited to References [1, 5, 25, 36, 37, 134, 135, 162, 166, 232].”
BackgroundInfluentialLin, T., Wang, Y., Liu, X., & Qiu, X. (2022) · AI Open · doi:10.1016/j.aiopen.2022.10.001 · Cited by 1,422
Lin, T., Wang, Y., Liu, X., & Qiu, X. (2021) · arXiv · doi:10.48550/arxiv.2106.04554 · Cited by 1,420
Explaining nonlinear classification decisions with deep taylor decomposition
Montavon, G., Lapuschkin, S., Binder, A., Samek, W., & Müller, K.-R. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2016.11.008 · Cited by 1,357
Detecting large-scale system problems by mining console logs
Xu, W., Huang, L., Fox, A., Patterson, D., & Jordan, M. I. (2009) · Proceedings of the ACM SIGOPS 22nd symposium on Operating systems principles · doi:10.1145/1629575.1629587 · Cited by 1,332 · cited 1×
“[223] propose an approach to detect and explain system problems by mining console logs.”
BackgroundDeep Learning for Anomaly Detection: A Survey
Chalapathy, R., & Chawla, S. (2019) · arXiv · doi:10.48550/arxiv.1901.03407 · Cited by 1,240 · cited 10×
“Since the seminal work in [105], anomaly detection has been well studied and there exists a plethora of comprehensive surveys and reviews on it, including but not limited to References [1, 5, 25, 36, 37, 134, 135, 162, 166, 232].”
BackgroundInfluentialEfficient algorithms for mining outliers from large data sets
Ramaswamy, S., Rastogi, R., & Shim, K. (2000) · Proceedings of the 2000 ACM SIGMOD international conference on Management of data · doi:10.1145/342009.335437 · Cited by 1,171 · cited 2×
“For example, KNN [170] defines objects with ‘far’ 𝑘-nearest neighbours as anomalies, LOF [26] treats objects with a low local density as anomalies, and Isolation Forest [121] considers ‘easily isolated’ objects as anomalies.”
Explainable artificial intelligence: A survey
Dosilovic, F. K., Brcic, M., & Hlupic, N. (2018) · 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO) · doi:10.23919/mipro.2018.8400040 · Cited by 1,147 · cited 21×
“After carefully scrutinizing existing surveys on XAI [13, 20, 30, 34, 62, 71, 120], we found that some criteria are often used to categorize existing XAI techniques.”
MethodInfluentialUnmasking Clever Hans predictors and assessing what machines really learn
Lapuschkin, S., Wäldchen, S., Binder, A., Montavon, G., Samek, W., & Müller, K.-R. (2019) · Nature Communications · doi:10.1038/s41467-019-08987-4 · Cited by 1,047 · cited 2×
“, to avoid the Clever Hans Phenomenon [112] thatÐin this contextÐrefers to anomaly detection models utilising spurious correlations and patterns in the data to”
MethodA Survey on the Explainability of Supervised Machine Learning
Burkart, N., & Huber, M. F. (2021) · Journal of Artificial Intelligence Research · doi:10.1613/jair.1.12228 · Cited by 1,008 · cited 20×
“Meanwhile, a refined taxonomy, distilled from existing surveys on XAI techniques, is presented below and used to categorize XAD methods.”
MethodInfluentialOutlier Detection for Temporal Data: A Survey
Gupta, M., Gao, J., Aggarwal, C. C., & Han, J. (2014) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2013.184 · Cited by 1,001 · cited 1×
“1 Related Work Compared to the abundance of taxonomies of anomaly detection methods, including but not limited to these surveys throughout the years [3, 36, 37, 78, 86, 161], the categorization of anomaly explanation methods involving XAD techniques has received relatively little attention so far [163, 186, 190, 217,…”
BackgroundA Unifying Review of Deep and Shallow Anomaly Detection
Ruff, L., Kauffmann, J. R., Vandermeulen, R. A., Montavon, G., Samek, W., Kloft, M., Dietterich, T. G., & Muller, K.-R. (2021) · Proceedings of the IEEE · doi:10.1109/jproc.2021.3052449 · Cited by 914 · cited 2×
“Moreover, it may also be known as an unusual, irregular, atypical, inconsistent, unexpected, rare, erroneous, faulty, fraudulent, malicious, unnatural, or strange object [182].”
BackgroundFooling LIME and SHAP: Adversarial Attacks on Post hoc Explanation Methods
Slack, D., Hilgard, S., Jia, E., Singh, S., & Lakkaraju, H. (2020) · Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society · doi:10.1145/3375627.3375830 · Cited by 900 · cited 1×
“In particular, post-hoc XAI techniques such as LIME and SHAP are easily manipulated [199].”
MethodA survey on unsupervised outlier detection in high‐dimensional numerical data
Zimek, A., Schubert, E., & Kriegel, H. (2012) · Statistical Analysis and Data Mining The ASA Data Science Journal · doi:10.1002/sam.11161 · Cited by 864
Novelty detection: a review—part 2:
Markou, M., & Singh, S. (2003) · Signal Processing · doi:10.1016/j.sigpro.2003.07.019 · Cited by 817 · cited 9×
“Since the seminal work in [105], anomaly detection has been well studied and there exists a plethora of comprehensive surveys and reviews on it, including but not limited to References [1, 5, 25, 36, 37, 134, 135, 162, 166, 232].”
BackgroundInfluentialA Modified Principal Component Technique Based on the LASSO
Jolliffe, I. T., Trendafilov, N. T., & Uddin, M. (2003) · Journal of Computational and Graphical Statistics · doi:10.1198/1061860032148 · Cited by 809 · cited 12×
“Given a dataset with 𝐷 features, they first compute and order the principal components using sparse PCA [93].”
MethodInfluentialvon Rueden, L., Mayer, S., Beckh, K., Georgiev, B., Giesselbach, S., Heese, R., Kirsch, B., Walczak, M., Pfrommer, J., Pick, A., et al. (2021) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2021.3079836 · Cited by 793 · cited 2×
“This paradigm of integrating prior knowledge into machine learning is called Informed Machine Learning [213], which has received increasing attention over the past few years.”
HOT SAX: Efficiently Finding the Most Unusual Time Series Subsequence
Keogh, E., Lin, J., & Fu, A. (2006) · Fifth IEEE International Conference on Data Mining (ICDM'05) · doi:10.1109/icdm.2005.79 · Cited by 696 · cited 2×
“HOT-SAX [101] and RRA [192], to discovery anomalies, respectively.”
BackgroundAutoencodersPreprint
Bank, D., Koenigstein, N., & Giryes, R. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2003.05991 · Cited by 673
Satisfiability modulo theories: introduction and applications
De Moura, L., & Bjørner, N. (2011) · Communications of the ACM · doi:10.1145/1995376.1995394 · Cited by 645
A Survey of Methods for Explaining Black Box Models
Guidotti, R., Monreale, A., Ruggieri, S., Turini, F., Giannotti, F., & Pedreschi, D. (2019) · ACM Computing Surveys · doi:10.1145/3236009 · Cited by 629 · cited 3×
“For instance, references [20, 76] analyze the XAI literature and propose important properties that should be considered when designing an XAI technique.”
BackgroundEvaluating the quality of machine learning explanations: a survey on methods and metrics
Zhou, J., Gandomi, A. H., Chen, F., & Holzinger, A. (2021) · Electronics · doi:10.3390/electronics10050593 · Cited by 606 · cited 1×
“Reference [224] performs a survey on the quality evaluation of machine learning explanations.”
BackgroundSurvey on Anomaly Detection using Data Mining Techniques
Agrawal, S., & Agrawal, J. (2015) · Procedia Computer Science · doi:10.1016/j.procs.2015.08.220 · Cited by 595 · cited 1×
“throughout the years [3, 35, 36, 77, 85, 158], the categorization of anomaly explanation methods involving XAD techniques has received relatively little attention so far [160, 183, 187, 211, 220].”
BackgroundAggarwal, C. C. (2017) · Springer eBooks · doi:10.1007/978-3-319-47578-3 · Cited by 566
Recent Advances in Recurrent Neural Networks
Salehinejad, H., Sankar, S., Barfett, J., Colak, E., & Valaee, S. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1801.01078 · Cited by 562
Addendum: Regularization and Variable Selection Via the Elastic Net
Zou, H., & Hastie, T. (2005) · Journal of the Royal Statistical Society Series B: Statistical Methodology · doi:10.1111/j.1467-9868.2005.00527.x · Cited by 559
A survey of anomaly detection techniques in financial domain
Ahmed, M., Mahmood, A. N., & Islam, M. R. (2016) · Future Generation Computer Systems · doi:10.1016/j.future.2015.01.001 · Cited by 528
Failure diagnosis using decision trees
Chen, M. Y., Zheng, A. X., Lloyd, J., et al. (2004) · International Conference on Autonomic Computing, 2004. Proceedings. · doi:10.1109/icac.2004.31 · Cited by 503
Explaining anomalies detected by autoencoders using Shapley Additive Explanations
Antwarg, L., Miller, R. M., Shapira, B., & Rokach, L. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.115736 · Cited by 496
Li, Z., Zhao, Y., Han, J., Su, Y., Jiao, R., Wen, X., & Pei, D. (2021) · Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining · doi:10.1145/3447548.3467075 · Cited by 464 · cited 1×
“[117] introduce InterFusion, a model based on hierarchical Variational AutoEncoder (HVAE) and Markov Chain Monte Carlo (MCMC) for detecting and explaining anomalies in multivariate time series data.”
MethodSelf-citationFailure diagnosis using decision trees
Chen, M., Zheng, A. X., Lloyd, J., Jordan, M. I., & Brewer, E. (2004) · International Conference on Autonomic Computing, 2004. Proceedings. · doi:10.1109/icac.2004.1301345 · Cited by 433
HiCS: High Contrast Subspaces for Density-Based Outlier Ranking
Keller, F., Muller, E., & Bohm, K. (2012) · 2012 IEEE 28th International Conference on Data Engineering · doi:10.1109/icde.2012.88 · Cited by 373 · cited 1×
“It can be seen that both methods [99, 149] decouple subspace search and anomaly scoring.”
MethodLoda: Lightweight on-line detector of anomalies
Pevný, T. (2016) · Machine Learning · doi:10.1007/s10994-015-5521-0 · Cited by 357 · cited 1×
“Additionally, Pevnỳ [170] presents Loda, an online anomaly detection model that can also provide explanations.”
BackgroundDiscovering spatio-temporal causal interactions in traffic data streams
Liu, W., Zheng, Y., Chawla, S., Yuan, J., & Xing, X. (2011) · Proceedings of the 17th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/2020408.2020571 · Cited by 350
Outlier Detection: Methods, Models, and Classification
Boukerche, A., Zheng, L., & Alfandi, O. (2021) · ACM Computing Surveys · doi:10.1145/3381028 · Cited by 334
InterpretML: A Unified Framework for Machine Learning Interpretability
Nori, H., Jenkins, S., Koch, P., & Caruana, R. (2019) · arXiv · doi:10.48550/arxiv.1909.09223 · Cited by 322 · cited 2×
“Second, they utilize two types of surrogate models to explain anomalies, including black-box anomaly detection models with a post-hoc local explanation (XGBoost [43] and LightGBM [97] with LIME or SHAP), and transparent anomaly detection models (ElasticNet [227] and EBM [150]).”
MethodAggarwal, C. C. (2015) · Data Mining · doi:10.1007/978-3-319-14142-8_8 · Cited by 305
Joint Detection and Recounting of Abnormal Events by Learning Deep Generic Knowledge
Hinami, R., Mei, T., & Satoh, S. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.391 · Cited by 268 · cited 1×
“[85] utilise a general CNN model and context-sensitive anomaly detectors to identify and explain abnormal events in ilms.”
MethodAttention Guided Anomaly Localization in Images
Venkataramanan, S., Peng, K.-C., Singh, R. V., & Mahalanobis, A. (2020) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-58520-4_29 · Cited by 247
IoT Healthcare Analytics: The Importance of Anomaly Detection
Ukil, A., Bandyoapdhyay, S., Puri, C., & Pal, A. (2016) · 2016 IEEE 30th International Conference on Advanced Information Networking and Applications (AINA) · doi:10.1109/aina.2016.158 · Cited by 222 · cited 1×
“For example, anomaly detection algorithms are being used to diagnose diseases in healthcare [213].”
Credit card fraud and detection techniques : a review
Delamaire, L., Abdou, H. A., & Pointon, J. (2009) · University of Salford Institutional Repository (University of Salford) · Cited by 219
Polya Trees and Random Distributions
Mauldin, R. D., Sudderth, W. D., & Williams, S. C. (1992) · The Annals of Statistics · doi:10.1214/aos/1176348766 · Cited by 216 · cited 3×
“Specifically, the Mondrian Process [177] is a family of hierarchical binary partitions of data and the Pólya Tree [136] is a non-parametric approach that can estimate the density function of binary partitions.”
MethodRecurrent Neural Network Attention Mechanisms for Interpretable System Log Anomaly Detection
Brown, A., Tuor, A., Hutchinson, B., & Nichols, N. (2018) · Proceedings of the First Workshop on Machine Learning for Computing Systems · doi:10.1145/3217871.3217872 · Cited by 198 · cited 2×
“Examples include Reference [28] and Reference [216].”
BackgroundFP-outlier: Frequent pattern based outlier detection
He, Z., Xu, X., Huang, Z., & Deng, S. (2005) · Computer Science and Information Systems · doi:10.2298/csis0501103h · Cited by 188 · cited 11×
“Examples include Rule-based models [83].”
BackgroundInfluentialContrastive explanation: a structural-model approach
Miller, T. (2021) · The Knowledge Engineering Review · doi:10.1017/s0269888921000102 · Cited by 186 · cited 1×
“Instead, we refer to recent papers for perspectives from HCI [203] and social science [142, 143], and leave a broader discussion of these aspects to a future article.”
BackgroundExplainable Deep One-Class Classification
Liznerski, P., Ruff, L., Vandermeulen, R. A., Franks, B. J., Kloft, M., & Müller, K.-R. (2020) · arXiv · doi:10.48550/arxiv.2007.01760 · Cited by 185
DAEMON: Unsupervised Anomaly Detection and Interpretation for Multivariate Time Series
Chen, X., Deng, L., Huang, F., Zhang, C., Zhang, Z., Zhao, Y. L., & Zheng, K. (2021) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde51399.2021.00228 · Cited by 183 · cited 1×
“[45] put forward DAEMON, which trains an Adversarial AutoEncoder (AAE) to learn the typical pattern of multivariate time series, and then use the reconstruction error to identify and explain anomalies.”
MethodToward Deep Supervised Anomaly Detection: Reinforcement Learning from Partially Labeled Anomaly Data
Pang, G., van den Hengel, A., Shen, C., & Cao, L. (2021) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3447548.3467417 · Cited by 179
Anomaly Detection in Autonomous Driving: A Survey
Bogdoll, D., Nitsche, M., & Zollner, J. M. (2022) · 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw56347.2022.00495 · Cited by 176 · cited 1×
“In addition, the self-driving car manufacturing industry applies anomaly detection algorithms on camera data to detect corner cases [23].”
MethodToward Explainable Deep Neural Network Based Anomaly Detection
Amarasinghe, K., Kenney, K., & Manic, M. (2018) · Conference on Human System Interactions · doi:10.1109/hsi.2018.8430788 · Cited by 168 · cited 1×
“[8] propose a framework for explainable Deep Neural Network (DNN) based anomaly detection.”
BackgroundPerformance of CUSUM Control Schemes for Serially Correlated Observations
Yashchin, E. (1993) · Technometrics · doi:10.2307/1269288 · Cited by 166 · cited 1×
“On this basis, they extend the CUMSUM algorithm [225] to detect anomalies by using the likelihood ratio between two windows of prediction residuals.”
MethodGEE: A Gradient-based Explainable Variational Autoencoder for Network Anomaly Detection
Nguyen, Q. P., Lim, K. W., Divakaran, D. M., Low, K. H., & Chan, M. C. (2019) · 2019 IEEE Conference on Communications and Network Security (CNS) · doi:10.1109/cns.2019.8802833 · Cited by 165 · cited 1×
“[151] introduce a framework called GEE to detect and explain anomalies in network traic.”
MethodSiebes, A., Vreeken, J., & Leeuwen, M. V. (2006) · Proceedings of the 2006 SIAM International Conference on Data Mining · doi:10.1137/1.9781611972764.35 · Cited by 165 · cited 1×
“They utilise the Krimp algorithm [196] as the compressor, which is trained on exclusively normal samples to capture normal behaviours.”
MethodSelf-citationA comprehensive survey of numeric and symbolic outlier mining techniques
Agyemang, M., Barker, K., & Alhajj, R. (2006) · Intelligent Data Analysis · doi:10.3233/ida-2006-10604 · Cited by 164 · cited 10×
“Since the seminal work in [105], anomaly detection has been well studied and there exists a plethora of comprehensive surveys and reviews on it, including but not limited to References [1, 5, 25, 36, 37, 134, 135, 162, 166, 232].”
BackgroundInfluentialAn improved k-prototypes clustering algorithm for mixed numeric and categorical data
Ji, J., Bai, T., Zhou, C., Ma, C., & Wang, Z. (2013) · Neurocomputing · doi:10.1016/j.neucom.2013.04.011 · Cited by 163 · cited 1×
“Second, they use an off-the-shelf clustering method such as K-Prototypes [93] to iteratively divide the non-anomalous instances into different regions until no anomalies are contained in the generated regions.”
MethodNeural Dynamics of Category Learning and Recognition: Attention, Memory Consolidation, and Amnesia
Carpenter, G. A., & Grossberg, S. (1987) · Advances in psychology · doi:10.1016/s0166-4115(08)60911-5 · Cited by 146 · cited 3×
“By virtue of the good properties of ART, one can obtain the feature differences between every two clusters, resulting in explanations for the anomalous instances.”
BackgroundInfluentialExplainable Deep Few-shot Anomaly Detection with Deviation NetworksPreprint
Pang, G., Ding, C., Shen, C., & van den Hengel, A. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2108.00462 · Cited by 141
Outlier Detection in Arbitrarily Oriented Subspaces
Kriegel, H., Kröger, P., Schubert, E., & Zimek, A. (2012) · Proceedings · doi:10.1109/icdm.2012.21 · Cited by 139 · cited 1×
“[109] irst project these data instances with all � features into subspaces of varying size, where the subspace is spanned by the � largest principle components using robust PCA.”
Automated Identification of Failure Causes in System Logs
Mariani, L., & Pastore, F. (2008) · Proceedings/Proceedings - International Symposium on Software Reliability Engineering · doi:10.1109/issre.2008.48 · Cited by 130 · cited 3×
“On this basis, they present the so-called Automata Violation Analyzer (AVA) to automatically explain anomalies detected by LFA.”
BackgroundStatistical selection of relevant subspace projections for outlier ranking
Muller, E., Schiffer, M., & Seidl, T. (2011) · 2011 IEEE 27th International Conference on Data Engineering · doi:10.1109/icde.2011.5767916 · Cited by 126 · cited 1×
“It can be seen that both methods [99, 149] decouple subspace search and anomaly scoring.”
MethodContext-Aware Activity Recognition and Anomaly Detection in Video
Zhu, Y., Nayak, N. M., & Roy-Chowdhury, A. K. (2013) · IEEE Journal of Selected Topics in Signal Processing · doi:10.1109/jstsp.2012.2234722 · Cited by 121 · cited 1×
“[231] propose a model to capture frequent motion and background patterns of activities in video data, treating patterns that deviate from learned frequent patterns as anomalies.”
BackgroundSelf-citationInterpretable Anomaly Detection with DIFFI: Depth-based feature importance of Isolation Forest
Carletti, M., Terzi, M., & Susto, G. A. (2023) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105730 · Cited by 118
Towards explaining anomalies: A deep Taylor decomposition of one-class models
Kauffmann, J., Müller, K., & Montavon, G. (2020) · Pattern Recognition · doi:10.1016/j.patcog.2020.107198 · Cited by 113 · cited 8×
“Examples include Layer-wise Relevance Propagation [95, 157, 194].”
MethodInfluentialAnomaly Detection using Graph Neural Networks
Chaudhary, A., Mittal, H., & Arora, A. (2019) · 2019 International Conference on Machine Learning, Big Data, Cloud and Parallel Computing (COMITCon) · doi:10.1109/comitcon.2019.8862186 · Cited by 106 · cited 2×
“Discussion: In addition to the above mentioned DNNs, namely AEs, RNNs, CNNs, GANs, Deep OCSVM and DevNet, other DNNs such as Graph Neural Networks [39] and Transformers [119] have become prevalent in anomaly detection.”
BackgroundHeart Sound Segmentation Using Bidirectional LSTMs With Attention
Fernando, T., Ghaemmaghami, H., Denman, S., Sridharan, S., Hussain, N., & Fookes, C. (2019) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2019.2949516 · Cited by 105 · cited 1×
“First, they apply the model proposed by Reference [68] to perform segmentation.”
MethodA Robust Interpretable Deep Learning Classifier for Heart Anomaly Detection Without Segmentation
Dissanayake, T., Fernando, T., Denman, S., Sridharan, S., Ghaemmaghami, H., & Fookes, C. (2021) · IEEE Journal of Biomedical and Health Informatics · doi:10.1109/jbhi.2020.3027910 · Cited by 105
Explaining Outliers by Subspace Separability
Micenkova, B., Ng, R. T., Dang, X.-H., & Assent, I. (2013) · 2013 IEEE 13th International Conference on Data Mining · doi:10.1109/icdm.2013.132 · Cited by 104 · cited 2×
“[141] criticise that “almost all existing algorithms stop at the point of providing anomaly ranking and leave the user without any explanation of why some data points deviate and how.”
BackgroundMethodRoshan, K., & Zafar, A. (2021) · International journal of Computer Networks & Communications · doi:10.5121/ijcnc.2021.13607 · Cited by 104 · cited 1×
“Meanwhile, Roshan & Zafar [179] develop an AutoEncoder (AE) based model incorporating the SHAP technique to detect and explain anomalies in computer network data.”
MethodPrinciples and Practice of Explainable Machine Learning
Belle, V., & Papantonis, I. (2021) · Frontiers in Big Data · doi:10.3389/fdata.2021.688969 · Cited by 99 · cited 27×
“Meanwhile, a refined taxonomy, distilled from existing surveys on XAI techniques, is presented below and used to categorize XAD methods.”
BackgroundMethodInfluentialDiscriminative features for identifying and interpreting outliers
Dang, X. H., Assent, I., Ng, R. T., Zimek, A., & Schubert, E. (2014) · IEEE International Conference on Data Engineering (ICDE) · doi:10.1109/icde.2014.6816642 · Cited by 97 · cited 1×
“[52] leverage spectral graph theory to achieve subspace anomaly detection.”
BackgroundDetecting outlying properties of exceptional objects
Angiulli, F., Fassetti, F., & Palopoli, L. (2009) · ACM Transactions on Database Systems · doi:10.1145/1508857.1508864 · Cited by 88 · cited 3×
“[11] extend their previous work [10] to explain a group of anomalies.”
BackgroundMethodHeart Sound Segmentation using Bidirectional LSTMs with Attention
Warnakulasuriya, T., Ghaemmaghami, H., Denman, S., et al. (2020) · QUT ePrints (Queensland University of Technology) · Cited by 88
An explainable and efficient deep learning framework for video anomaly detection
Wu, C., Shao, S., Tunc, C., Satam, P., & Hariri, S. (2022) · Cluster Computing · doi:10.1007/s10586-021-03439-5 · Cited by 88 · cited 3×
“Ref Spec Pers Tech Data Loc Pros Cons [156] A F Feature selection Static TC G Handles noisy features well Only applicable to categorical data [158] A F Feature selection Static TC G Linear time complexity to data size Only applicable to categorical data [82] A F & S Feature selection + Instance selection Static TC G J…”
BackgroundMethodExplainable Anomaly Detection Framework for Maritime Main Engine Sensor Data
Kim, D., Antariksa, G., Handayani, M. P., Lee, S.-B., & Lee, J. (2021) · Sensors · doi:10.3390/s21155200 · Cited by 84 · cited 1×
“[103] utilize Isolation Forest on sensor stream data of marine engines to keep track of unusual engine conditions.”
MethodDiscovering outlying aspects in large datasets
Vinh, N. X., Chan, J., Romano, S., Bailey, J., Leckie, C., Ramamohanarao, K., & Pei, J. (2016) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-016-0453-2 · Cited by 83 · cited 4×
“1 Related Work Compared to the abundance of taxonomies of anomaly detection methods, including but not limited to these surveys throughout the years [3, 36, 37, 78, 86, 161], the categorization of anomaly explanation methods involving XAD techniques has received relatively little attention so far [163, 186, 190, 217,…”
BackgroundMethodInfluentialSerradilla, O., Zugasti, E., Ramirez de Okariz, J., Rodriguez, J., & Zurutuza, U. (2021) · Applied Sciences · doi:10.3390/app11167376 · Cited by 83 · cited 1×
“[193] utilise different machine learning approaches to detect, predict and explain anomalies in press machines to achieve predictable maintenance.”
MethodE-SFD: Explainable Sensor Fault Detection in the ICS Anomaly Detection System
Hwang, C., & Lee, T. (2021) · IEEE Access · doi:10.1109/access.2021.3119573 · Cited by 82 · cited 3×
“Moreover, Hwang & Lee [87] propose a bidirectional stackable LSTM-based anomaly detection model for industrial control system anomaly detection.”
BackgroundInfluentialLangone, R., Cuzzocrea, A., & Skantzos, N. (2020) · Data & Knowledge Engineering · doi:10.1016/j.datak.2020.101850 · Cited by 81 · cited 2×
“Reference [110] leverages regularized Logistic Regression to identify anomalies in time series.”
BackgroundMethodTime series anomaly discovery with grammar-based compression
Senin, P., Lin, J., Wang, X., Oates, T., Gandhi, S., Boedihardjo, A. P., Chen, C., & Frankenstein, S. (2015) · Movebank · doi:10.5441/002/edbt.2015.42 · Cited by 80
FRaC: a feature-modeling approach for semi-supervised and unsupervised anomaly detection
Noto, K., Brodley, C., & Slonim, D. (2012) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-011-0234-x · Cited by 79
Robust and Rapid Clustering of KPIs for Large-Scale Anomaly Detection
Li, Z., Zhao, Y., Liu, R., & Pei, D. (2018) · International Workshop on Quality of Service · doi:10.1109/iwqos.2018.8624168 · Cited by 78 · cited 1×
“For example, large internet companies often develop an anomaly detection system to monitor a large number of key performance indicators, aiming to ensure the reliability of their service platform [118].”
BackgroundSelf-citationBarbado, A., Corcho, Ó., & Benjamins, R. (2022) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.116100 · Cited by 77 · cited 1×
“Next, Reference [18] defines some criteria to evaluate”
BackgroundPsychological foundations of explainability and interpretability in artificial intelligence
Broniatowski, D. A. (2021) · doi:10.6028/nist.ir.8367 · Cited by 76 · cited 2×
“Note that some researchers [27, 145, 198] distinguish between the terms ‘interpretation’ and ‘explanation’, the terms ‘interpretable’ and ‘explainable’, and the terms ‘explainability’ and ‘interpretability’.”
BackgroundContextual Outlier Interpretation
Liu, N., Shin, D., & Hu, X. (2018) · Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2018/341 · Cited by 74 · cited 2×
“On the contrary, model-agnostic techniques such as Reference [123] usually have an Explanation-Deinition that may difer from the Detection-Deinition.”
MethodAguilar, D. L., Medina‐Pérez, M. A., Loyola‐González, O., Choo, K. R., & Bucheli-Susarrey, E. (2022) · IEEE Transactions on Dependable and Secure Computing · doi:10.1109/tdsc.2022.3148331 · Cited by 72 · cited 1×
“[4] propose the Decision Tree-based AutoEncoder (DTAE) model to detect anomalies.”
BackgroundSAIFE: Unsupervised Wireless Spectrum Anomaly Detection with Interpretable Features
Rajendran, S., Meert, W., Lenders, V., & Pollin, S. (2018) · IEEE International Symposium on Dynamic Spectrum Access Networks · doi:10.1109/dyspan.2018.8610471 · Cited by 72 · cited 1×
“[172] present an AAE based anomaly detection method named SAIFE.”
MethodMining outlying aspects on numeric data
Duan, L., Tang, G., Pei, J., Bailey, J., Campbell, A., & Tang, C. (2015) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-014-0398-2 · Cited by 70 · cited 1×
“[64] develop a method to find its minimal outlying subspace, i.”
MethodThe Odd One Out: Identifying and Characterising Anomalies
Smets, K., & Vreeken, J. (2011) · SIAM International Conference on Data Mining · doi:10.1137/1.9781611972818.69 · Cited by 69 · cited 2×
“More concretely, techniques such as Probabilistic Models (e.g., Mondrian Pólya Forest and other distribution or density estimation based approaches), Data Cube structure, Incremental Clustering, MDL based Pattern Compression, Region-Partition trees are harnessed to detect and explain anomalies in tabular data.”
A survey on outlier explanations
Panjei, E., Gruenwald, L., Leal, E., Nguyen, C., & Silvia, S. (2022) · The VLDB Journal · doi:10.1007/s00778-021-00721-1 · Cited by 67 · cited 5×
“[163] particularly treat anomaly ranking as a subcategory of anomaly explanation methods.”
BackgroundInfluentialImproving Spacecraft Health Monitoring with Automatic Anomaly Detection Techniques
Fuertes, S., Picart, G., Tourneret, J., Chaâri, L., Ferrari, A., & Richard, C. (2016) · SpaceOps Conference (SpaceOps) · doi:10.2514/6.2016-2430 · Cited by 66 · cited 1×
“In other safety-critical areas—such as spacecraft design—anomaly detection algorithms are used to detect sensor faults [69].”
BackgroundTree Space Prototypes: Another Look at Making Tree Ensembles Interpretable
Tan, S., Soloviev, M., Hooker, G., & Wells, M. T. (2020) · ACM-IMS Foundations of Data Science Conference (FODS) · doi:10.1145/3412815.3416893 · Cited by 64 · cited 9×
“Moreover, model-specific techniques such as References [7, 31, 94] generally have consistent Explanation-Definition and Detection-Definition of anomaly as they explore the internal structure of anomaly detection model to generate explanations.”
BackgroundInfluentialShapley Values of Reconstruction Errors of PCA for Explaining Anomaly Detection
Takeishi, N. (2019) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw.2019.00117 · Cited by 63 · cited 1×
“However, Takeishi [210] argues that by simply looking at the reconstruction error of each feature, one may fail to ind the true cause of the anomaly.”
BackgroundA decomposition of the outlier detection problem into a set of supervised learning problems
Paulheim, H., & Meusel, R. (2015) · Machine Learning · doi:10.1007/s10994-015-5507-y · Cited by 62 · cited 2×
“Reference [151] and Reference [165] try to find a relevant feature subset for anomaly detection by exploring the correlations between features.”
BackgroundMathonsi, T., & van Zyl, T. L. (2021) · Forecasting · doi:10.3390/forecast4010001 · Cited by 60 · cited 1×
“Additionally, Mathonsi & van Zyl [136] present Multivariate Exponential Smoothing Long Short-Term Memory (MES-LSTM) that combines statistics and deep learning.”
BackgroundPang, G., Cao, L., Chen, L., Lian, D., & Liu, H. (2018) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v32i1.11692 · Cited by 59 · cited 1×
“[157] propose an embedded feature selection method for anomaly detection, dubbed CINFO, which is an ensemble of sequential ensemble learners.”
MethodBeyond Outlier Detection: Outlier Interpretation by Attention-Guided Triplet Deviation Network
Xu, H., Wang, Y., Jian, S., Huang, Z., Wang, Y., Liu, N., & Li, F. (2021) · International World Wide Web Conference (WWW) · doi:10.1145/3442381.3449868 · Cited by 57 · cited 1×
“[222] propose Attention-guided Triplet deviation network for Outlier interpretatioN (ATON) to explain anomalies in a post-hoc fashion.”
Beyond Outlier Detection: LookOut for Pictorial Explanation
Gupta, N., Eswaran, D., Shah, N., Akoglu, L., & Faloutsos, C. (2019) · Lecture notes in computer science · doi:10.1007/978-3-030-10925-7_8 · Cited by 57
Local Outlier Detection with Interpretation
Dang, X. H., Micenková, B., Assent, I., & Ng, R. T. (2013) · Lecture Notes in Computer Science · doi:10.1007/978-3-642-40994-3_20 · Cited by 56 · cited 10×
“Moreover, model-specific techniques such as References [7, 31, 94] generally have consistent Explanation-Definition and Detection-Definition of anomaly as they explore the internal structure of anomaly detection model to generate explanations.”
BackgroundInfluentialAnomaly Detection in Asset Degradation Process Using Variational Autoencoder and Explanations
Jakubowski, J., Stanisz, P., Bobek, S., & Nalepa, G. J. (2021) · Sensors · doi:10.3390/s22010291 · Cited by 56 · cited 1×
“[91] propose a Variational AutoEncoder (VAE) model combined with Shapley values to detect and interpret anomalies in an asset degradation process.”
MethodExplaining anomalies in groups with characterizing subspace rules
Macha, M., & Akoglu, L. (2018) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-018-0585-7 · Cited by 54 · cited 1×
“Furthermore, Macha & Akoglu [129] propose x-PACS to explain anomalies in a group in three steps.”
BackgroundLocally Interpretable One-Class Anomaly Detection for Credit Card Fraud Detection
Wu, T.-Y., & Wang, Y.-T. (2021) · Conference on Technologies and Applications of Artificial Intelligence (TAAI) · doi:10.1109/taai54685.2021.00014 · Cited by 53 · cited 1×
“For example, Wu & Wang [221] propose a neural network based model incorporating LIME techniques to detect and”
MethodYepmo, V., Smits, G., & Pivert, O. (2022) · Data & Knowledge Engineering · doi:10.1016/j.datak.2021.101946 · Cited by 52 · cited 4×
“Further, although Explanations by Feature Importance and Explanations by Feature Values mainly differ in the granularity of provided explanations, Reference [220] regards them as two distinct categories.”
BackgroundMethodInfluentialBabenko, A., Mariani, L., & Pastore, F. (2009) · ACM SIGSOFT International Symposium on Software Testing and Analysis (ISSTA) · doi:10.1145/1572272.1572300 · Cited by 51 · cited 1×
“For instance, Babenko & Pastore [15] leverage LFA [131] to detect anomalies in system logs.”
BackgroundSequential Feature Explanations for Anomaly Detection
Siddiqui, M. A., Fern, A., Dietterich, T. G., & Wong, W.-K. (2019) · ACM Transactions on Knowledge Discovery from Data · doi:10.1145/3230666 · Cited by 50 · cited 1×
“Reference [191] points out that the effort required to investigate an anomaly is usually proportional to the number”
BackgroundBöhmer, K., & Rinderle‐Ma, S. (2019) · Information Systems · doi:10.1016/j.is.2019.101438 · Cited by 49 · cited 1×
“Moreover, Reference [24] develops the ADAR model to detect and explain anomalies in process runtime”
MethodPang, G., Cao, L., Chen, L., & Liu, H. (2017) · International Joint Conference on Artificial Intelligence · doi:10.24963/ijcai.2017/360 · Cited by 45 · cited 1×
“[159] introduce a wrapper feature selection framework for anomaly detection.”
MethodA one-class classification decision tree based on kernel density estimation
Itani, S., Lecron, F., & Fortemps, P. (2020) · Applied Soft Computing · doi:10.1016/j.asoc.2020.106250 · Cited by 45
The Clever Hans Effect in Anomaly DetectionPreprint
Kauffmann, J., Ruff, L., Montavon, G., & Müller, K. (2020) · arXiv (Cornell University) · doi:10.48550/arxiv.2006.10609 · Cited by 44
Pang, G., Cao, L., Chen, L., & Liu, H. (2016) · 2016 IEEE 16th International Conference on Data Mining (ICDM) · doi:10.1109/icdm.2016.0052 · Cited by 43 · cited 2×
“[158] propose two ilter-based unsupervised selection methods, namely CBRW_FS and CBRW, which select a subset of features independently from subsequent anomaly detection methods.”
BackgroundMethodDetecting Cyber Attacks Using Anomaly Detection with Explanations and Expert Feedback
Siddiqui, A., Stokes, J. W., Seifert, C., Argyle, E., McCann, R., Neil, J., & Carroll, J. (2019) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp.2019.8683212 · Cited by 43 · cited 1×
“[195] apply Isolation Forest [122] to detect cyber attacks and leverages SFEs to generate explanations.”
MethodExplainable anomaly detection for Hot-rolling industrial process
Jakubowski, J., Stanisz, P., Bobek, S., & Nalepa, G. J. (2021) · IEEE International Conference on Data Science and Advanced Analytics (DSAA) · doi:10.1109/dsaa53316.2021.9564228 · Cited by 37 · cited 1×
“[92] examine the issue of anomaly detection when hot rolling slabs into coils.”
BackgroundExplainable Anomaly Detection for District Heating Based on Shapley Additive Explanations
Park, S., Moon, J., & Hwang, E. (2020) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw51313.2020.00111 · Cited by 35 · cited 1×
“[165] employ SHAP [128] to explain anomalies by showing feature contributions.”
MethodStackVAE-G: An efficient and interpretable model for time series anomaly detection
Li, W., Hu, W., Chen, T., Chen, N., & Feng, C. (2022) · AI Open · doi:10.1016/j.aiopen.2022.07.001 · Cited by 35
Flexible and adaptive subspace search for outlier analysis
Keller, F., Müller, E., Wixler, A., & Böhm, K. (2013) · ACM International Conference on Information and Knowledge Management (CIKM) · doi:10.1145/2505515.2505560 · Cited by 34 · cited 1×
“[100] propose RefOut, which consists of three steps.”
BackgroundConvolutional Neural Networks for Unsupervised Anomaly Detection in Text Data
Gorokhov, O., Petrovskiy, M., & Mashechkin, I. (2017) · Lecture Notes in Computer Science · doi:10.1007/978-3-319-68935-7_54 · Cited by 34 · cited 28×
“Accordingly, a variety of CNN architectures have been proposed, including LeNet, AlexNet, GoogleNet, VGGNet, Inception V4, ResNet, etc.”
BackgroundMethodInfluentialBarbado, A., & Corcho, Ó. (2022) · Engineering Applications of Artificial Intelligence · doi:10.1016/j.engappai.2022.105222 · Cited by 34
Outlying property detection with numerical attributes
Angiulli, F., Fassetti, F., Manco, G., & Palopoli, L. (2017) · Data Mining and Knowledge Discovery · doi:10.1007/s10618-016-0458-x · Cited by 33
Harnessing Prior Knowledge for Explainable Machine Learning: An Overview
Beckh, K., Müller, S., Jakobs, M., Toborek, V., Tan, H., Fischer, R., Welke, P., Houben, S., & von Rueden, L. (2023) · IEEE Conference on Secure and Trustworthy Machine Learning (SaTML) · doi:10.1109/satml54575.2023.00038 · Cited by 33
Explainable outlier detection: What, for Whom and Why?
Sejr, J. H., & Schneider-Kamp, A. (2021) · Machine Learning with Applications · doi:10.1016/j.mlwa.2021.100172 · Cited by 32
EXAD: A System for Explainable Anomaly Detection on Big Data Traces
Song, F., Diao, Y., Read, J., Stiegler, A., & Bifet, A. (2018) · IEEE ... International Conference on Data Mining workshops · doi:10.1109/icdmw.2018.00204 · Cited by 31 · cited 1×
“[204] develop the EXAD system to identify and interpret anomalies from Apache Spark traces.”
MethodDiscovering Characterizations of the Behavior of Anomalous Subpopulations
Angiulli, F., Fassetti, F., & Palopoli, L. (2012) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2012.58 · Cited by 31 · cited 1×
“[11] extend their previous work [10] to explain a group of anomalies.”
BackgroundCheong, M.-S., Wu, M.-C., & Huang, S.-H. (2021) · IEEE Access · doi:10.1109/access.2021.3077067 · Cited by 30 · cited 1×
“[47] set up a SpatioTemporal Convolutional Neural Networkśbased Relational Network (STCNN-RN) to detect anomalous events in inancial markets.”
BackgroundA New Effective and Efficient Measure for Outlying Aspect Mining
Samariya, D., Aryal, S., Ting, K. M., & Ma, J. (2020) · Lecture notes in computer science · doi:10.1007/978-3-030-62008-0_32 · Cited by 30 · cited 1×
“[185] point out an issue of using Z-score normalisation of density to rank subspaces for outlying aspects mining.”
MethodExplainable AI: Using Shapley Value to Explain Complex Anomaly Detection ML-Based Systems
Zou, J., & Petrosian, O. (2020) · Frontiers in artificial intelligence and applications · doi:10.3233/faia200777 · Cited by 30
Multiclass Anomaly Detection in Flight Data Using Semi-Supervised Explainable Deep Learning Model
Memarzadeh, M., Matthews, B., & Templin, T. (2022) · Journal of Aerospace Information Systems · doi:10.2514/1.i010959 · Cited by 28
X-MAN: Explaining multiple sources of anomalies in video
Szymanowicz, S., Charles, J., & Cipolla, R. (2021) · IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW) · doi:10.1109/cvprw53098.2021.00360 · Cited by 27 · cited 1×
“[208] introduce a model to detect and explain anomalies in videos.”
BackgroundDetection and interpretation of anomalies in building energy use through inverse modeling
Gunay, B., Shen, W., Newsham, G. R., et al. (2019) · Science and Technology for the Built Environment · doi:10.1080/23744731.2019.1565550 · Cited by 27 · cited 1×
“[29], the heating and cooling load patterns of buildings are studied using three inverse models, including a univariate change point model, a regression trees based model, and an DNN based model.”
MethodAdditive Explanations for Anomalies Detected from Multivariate Temporal Data
Giurgiu, I., & Schumann, A. (2019) · Proceedings of the 28th ACM International Conference on Information and Knowledge Management · doi:10.1145/3357384.3358121 · Cited by 27 · cited 10×
“Moreover, model-specific techniques such as References [7, 31, 94] generally have consistent Explanation-Definition and Detection-Definition of anomaly as they explore the internal structure of anomaly detection model to generate explanations.”
BackgroundInfluentialDiscrete neural representations for explainable anomaly detection
Szymanowicz, S., Charles, J., & Cipolla, R. (2022) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv51458.2022.00157 · Cited by 26 · cited 1×
“[209] develop a method for detecting and automatically explaining anomalous events in video.”
MethodAnomaly Detection in Videos Using Two‐Stream Autoencoder with Post Hoc Interpretability
Feng, J., Liang, Y., & Li, L. (2021) · Computational Intelligence and Neuroscience · doi:10.1155/2021/7367870 · Cited by 25 · cited 1×
“[68] develop a Two-Stream AutoEncoder (AE) based model to detect abnormal events in videos and then utilise a Feature Map Visualization method to interpret the anomalies.”
Method(2004) · doi:10.1016/b978-012088469-8/50123-6 · Cited by 24
A general-purpose method for applying Explainable AI for Anomaly Detection
Sipple, J., & Youssef, A. (2022) · International Syposium on Methodologies for Intelligent Systems · doi:10.48550/arxiv.2207.11564 · Cited by 24
Toward Explainable Deep Anomaly Detection
Pang, G., & Aggarwal, C. (2021) · Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining · doi:10.1145/3447548.3470794 · Cited by 23 · cited 1×
“One natural question is, how can we trust these systems without understanding and validating the underlying rationale of the involved anomaly detection components? For this reason, XAD aims to not only provide accurate anomaly detection results, but also to provide tangible explanations of why a specific object is det…”
BackgroundCheng, X., Wang, Z., Yang, X., Xu, L., & Liu, Y. (2021) · Computers, Environment and Urban Systems · doi:10.1016/j.compenvurbsys.2021.101627 · Cited by 22 · cited 10×
“Moreover, model-specific techniques such as References [7, 31, 94] generally have consistent Explanation-Definition and Detection-Definition of anomaly as they explore the internal structure of anomaly detection model to generate explanations.”
BackgroundInfluentialVisual explanation of neural network based rotation machinery anomaly detection system
Saeki, M., Ogata, J., Murakawa, M., & Ogawa, T. (2019) · IEEE Conference on Prognostics and Health Management · doi:10.1109/icphm.2019.8819396 · Cited by 22 · cited 1×
“[183] present a CNN based method to detect and explain machinery faults based on vibration data.”
MethodExplainable LSTM Model for Anomaly Detection in HDFS Log File using Layerwise Relevance Propagation
Patil, A., Wadekar, A., Gupta, T., Vijan, R., & Kazi, F. (2019) · 2019 IEEE Bombay Section Signature Conference (IBSSC) · doi:10.1109/ibssc47189.2019.8973044 · Cited by 21 · cited 1×
“[167] use LSTM to detect anomalies in such systems.”
MethodEstimation of Dimensions Contributing to Detected Anomalies with Variational AutoencodersPreprint
Ikeda, Y., Tajiri, K., Nakano, Y., Watanabe, K., & Ishibashi, K. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1811.04576 · Cited by 20
Isolation-based feature Selection for Unsupervised Outlier Detection
Yang, Q., Singh, J., & Lee, J. (2019) · Annual Conference of the PHM Society · doi:10.36001/phmconf.2019.v11i1.824 · Cited by 19 · cited 2×
“The isolation forest based feature selection method, IBFS [218], simply selects features that contribute the most to the outlyingness of anomalies reported by the Isolation Forest method.”
MethodA General-Purpose Method for Applying Explainable AI for Anomaly Detection
Sipple, J., & Youssef, A. (2022) · Lecture Notes in Computer Science · doi:10.1007/978-3-031-16564-1_16 · Cited by 18
A Framework for Outlier Description Using Constraint Programming
Kuo, C.-T., & Davidson, I. (2016) · Proceedings of the AAAI Conference on Artificial Intelligence · doi:10.1609/aaai.v30i1.10174 · Cited by 18 · cited 1×
“Concretely, Kuo & Davidson [110] formally define the outlier description (namely anomaly explanation) problem and propose a Constraint Programming (CP) based framework to encode the problem.”
BackgroundVehicle operating state anomaly detection and results virtual reality interpretation
Alizadeh, M., Hamilton, M., Jones, P., Ma, J., & Jaradat, R. (2021) · Expert Systems with Applications · doi:10.1016/j.eswa.2021.114928 · Cited by 18 · cited 10×
“Moreover, model-specific techniques such as References [7, 32, 95] generally have consistent Explanation-Definition and Detection-Definition of anomaly as they explore the internal structure of an anomaly detection model to generate explanations.”
MethodInfluentialInterpretable Anomaly Detection in Event Sequences via Sequence Matching and Visual Comparison
Guo, S., Jin, Z., Chen, Q., Gotz, D., Hongyuan Zha, B., & Cao, N. (2021) · IEEE Transactions on Visualization and Computer Graphics · doi:10.1109/tvcg.2021.3093585 · Cited by 18 · cited 1×
“[77] set up a Sequence-to-Sequence VAE based model to detect anomalies in event sequences.”
MethodRamirez, E., Wimmer, M., & Atzmueller, M. (2019) · Lecture Notes in Computer Science · doi:10.1007/978-3-030-37446-4_11 · Cited by 17 · cited 3×
“[174] S P Symbolic representation using PAA Static MTS G Enables human-in-theloop Only applicable to symbolic based AD such as HOT-SAX and RRA [60] A F Pre-extracted feature representations (MFCCs/spectrogram); SHAP + Occlusion maps Heart sound signals/UTS L & G Simple, stable and efficient architecture Only applicabl…”
MethodInfluentialAnomaly Detection and Explanation in Context-Aware Software Product Lines
Mauro, J., Nieke, M., Seidl, C., et al. (2017) · doi:10.1145/3109729.3109752 · Cited by 16 · cited 1×
“Besides, Reference [137] proposes HyVarRec to detect and explain anomalous traces for context-aware software product lines.”
BackgroundInterpretableSAD: Interpretable Anomaly Detection in Sequential Log Data
Han, X., Cheng, H., Xu, D., & Yuan, S. (2021) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata52589.2021.9671642 · Cited by 16 · cited 1×
“[80] present InterpretableSAD, a Negative Sampling based method for detecting and interpreting anomalies in sequential log data.”
MethodData-Efficient and Interpretable Tabular Anomaly Detection
Chang, C.-H., Yoon, J., Arık, S. Ö., Udell, M. R., & Pfister, T. (2023) · ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD) · doi:10.1145/3580305.3599294 · Cited by 16 · cited 1×
“[38] present an explainable anomaly detection model named DIAD based on GAMs.”
MethodCounterexample Explanation by Anomaly Detection
Leue, S., & Befrouei, M. T. (2012) · Lecture notes in computer science · doi:10.1007/978-3-642-31759-0_5 · Cited by 15 · cited 1×
“In addition, Leue & Befrouei [113] design a method to explain counterexamples that are symptoms of deadlocks in concurrent systems.”
MethodScalable anomaly detection in manufacturing systems using an interpretable deep learning approach
Schlegl, T., Schlegl, S., West, N., & Deuse, J. (2021) · Procedia CIRP · doi:10.1016/j.procir.2021.11.261 · Cited by 15 · cited 2×
“Reference [185] constructs a deep neural network-based model that can learn interpretable feature representations from unlabeled time series, facilitating the evaluation and deployment of subsequent anomaly detection algorithms.”
BackgroundUse of Taxi-Trip Data in Analysis of Demand Patterns for Detection and Explanation of Anomalies
Markou, I., Rodrigues, F., & Pereira, F. C. (2017) · Transportation Research Record Journal of the Transportation Research Board · doi:10.3141/2643-15 · Cited by 14 · cited 1×
“[133] create a tool for exploiting internet data to find abnormalities in transportation networks and connecting them to unique events.”
MethodInterpretable Unsupervised Anomaly Detection For RAN Cell Trace Analysis
Chawla, A., Jacob, P., Feghhi, S., Rughwani, D., van der Meer, S., & Fallon, S. (2020) · 2020 16th International Conference on Network and Service Management (CNSM) · doi:10.23919/cnsm50824.2020.9269108 · Cited by 14 · cited 1×
“[40] set up a Sparse AutoEncoder (SAE) based anomaly detection algorithm and then applies kernel SHAP to explain the results.”
MethodToward Scalable and Unified Example-Based Explanation and Outlier Detection
Chong, P., Cheung, N.-M., Elovici, Y., & Binder, A. (2022) · IEEE Transactions on Image Processing · doi:10.1109/tip.2021.3127847 · Cited by 14 · cited 10×
“Moreover, model-specific techniques such as References [7, 31, 94] generally have consistent Explanation-Definition and Detection-Definition of anomaly as they explore the internal structure of anomaly detection model to generate explanations.”
BackgroundMethodInfluentialDas, S., Islam, M. R., Jayakodi, N. K., et al. (2019) · arXiv (Cornell University) · doi:10.48550/arxiv.1901.08930 · Cited by 14
Outlier Detection Applying an Innovative User Transaction Modeling with Automatic Explanation
Perez, D. G., & Lavalle, M. M. (2011) · Electronics, Robotics and Automotive Mechanics Conference · doi:10.1109/cerma.2011.14 · Cited by 13 · cited 1×
“Additionally, Perez & Lavalle [169] devise the alleged User Model to detect potential fraud in bank transactions, where they it manually selected features into a threshold-based rule model, classifying the model outputs in the form of fraud probability into ive categories.”
BackgroundAn explainable outlier detection method using region-partition trees
Park, C. H., & Kim, J. (2020) · The Journal of Supercomputing · doi:10.1007/s11227-020-03384-x · Cited by 12 · cited 1×
“Instead of using patterns to represent what is normal, Park & Kim [164] put forward a model that is an ensemble of Region-Partition (RP) trees.”
MethodAnoMili: Spoofing Prevention and Explainable Anomaly Detection for the 1553 Military Avionic BusPreprint
Levy, E., Maman, N., Shabtai, A., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2202.06870 · Cited by 12
Li, J., Zhang, J., Wang, J., et al. (2021) · IEEE International Conference on Big Data (BigData) · doi:10.1109/bigdata52589.2021.9671744 · Cited by 11 · cited 1×
“[115] develop a Variational AutoEncoder (VAE) and genetic algorithm (GA) based framework, called VAGA, to detect anomalies in high-dimensional data and search corresponding abnormal subspaces.”
Self-citationSelf-Supervised and Interpretable Anomaly Detection using Network TransformersPreprint
Marino, D., Wickramasinghe, C. S., Rieger, C. G., et al. (2022) · arXiv (Cornell University) · doi:10.48550/arxiv.2202.12997 · Cited by 10
Visualization and Explanation of Payload-Based Anomaly Detection
Rieck, K., & Laskov, P. (2009) · European Conference on Computer Network Defense (EC2ND) · doi:10.1109/ec2nd.2009.12 · Cited by 9 · cited 1×
“For example, Rieck & Laskov [177] propose a technique for explaining intrusion detection results.”
MethodSejr, J. H., Christiansen, T., Dvinge, N., Hougesen, D., Schneider–Kamp, P., & Zimek, A. (2021) · Applied Sciences · doi:10.3390/app11052270 · Cited by 9
Nor, A. K. M., Pedapati, S. R., & Muhammad, M. B. (2021) · Preprints.org · doi:10.20944/preprints202109.0034.v1 · Cited by 9 · cited 1×
“[152] present a probabilistic LSTM based model combined with SHAP to detect and interpret anomalies in gas turbines.”
MethodAn Interpretable Lane Change Detector Algorithm based on Deep Autoencoder Anomaly Detection
Candido, O. D., Binder, M., & Utschick, W. (2021) · IEEE Intelligent Vehicles Symposium · doi:10.1109/iv48863.2021.9575599 · Cited by 9 · cited 1×
“[56] present the Lane Change Detector (LCD) model to detect and explain when the surrounding vehicles of an ego vehicle change their lanes.”
BackgroundPANDA: Human-in-the-Loop Anomaly Detection and Explanation
Smits, G., Lesot, M., Yepmo, V., & Pivert, O. (2022) · Communications in computer and information science · doi:10.1007/978-3-031-08974-9_57 · Cited by 9 · cited 1×
“Instead, we refer to recent papers for perspectives from HCI [203] and social science [142, 143], and leave a broader discussion of these aspects to a future article.”
BackgroundAnomaly Detection and Interpretation using Multimodal Autoencoder and Sparse OptimizationPreprint
Ikeda, Y., Ishibashi, K., Nakano, Y., et al. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1812.07136 · Cited by 9
Research works printed but not linked: 22
- 10.5555/645920.672836 · doi:10.5555/645920.672836
- Xingyan Bin Ying Zhao and Bilong Shen. 2016. Abnormal subspace sparse PCA for anomaly detection and interpretation. In ODDx3’15:ACM SIGKDD 2015 Workshop . Association for Computing Machinery.
- Chandranil Chakraborttii and Heiner Litz. 2020. Explaining SSD failures using anomaly detection. In Proceedings of the Non-Volatile Memory Workshop. 1.
- N. F. Chen, Zhiyuan Du, and Khin Hua Ng. 2018. Scene graphs for interpretable video anomaly classification. In Proceedings of the Conference on Neural Information Processing Systems Workshop on Visually Grounded Interaction and Language.
- Ian Davidson. 2007. Anomaly detection, explanation and visualization. SGI, Tokyo, Japan, Tech. Rep (2007).
- Levent Ertoz, Eric Eilertson, Aleksandar Lazarevic, Pang-Ning Tan, Vipin Kumar, Jaideep Srivastava, and Paul Dokas. 2004. Minds-minnesota intrusion detection system. Next Generation Data Mining (2004), 199–218.
- Nico Gnoss Martin Schultz and Marina Tropmann-Frick. 2022. XAI in the audit domain-explaining an autoencoder model for anomaly detection. In 17th International Conference on Wirtschaftsinformatik . AISeL.
- Nirmal Sobha Kartha Clément Gautrais and Vincent Vercruyssen. 2021. Why are you weird? Infusing interpretability in isolation forest for anomaly detection. In Proceedings of the Explainable Agency in AI Workshop (AAAI’21) . 51–57.
- Guolin Ke Qi Meng Thomas Finley Taifeng Wang Wei Chen Weidong Ma Qiwei Ye and Tie-Yan Liu. 2017. Lightgbm: A highly efficient gradient boosting decision tree. In Proceedings of the 31st International Conference on Neural Information Processing Systems (Long Beach California USA) (NIPS’17) . Curran Associates Inc. 3149–3157.
- Edwin M. Knorr and Raymond T. Ng. 1998. Algorithms for mining distance-based outliers in large datasets. In Proceedings of the VLDB. Citeseer, 392–403.
- Edwin M. Knorr and Raymond T. Ng. 1999. Finding intensional knowledge of distance-based outliers. In Proceedings of the VLDB. Citeseer, 211–222.
- Martin Kopp, Tomáš Pevnỳ, and Martin Holena. 2014. Interpreting and clustering outliers with sapling random forests. In Proceedings of the European Conference on Information Technologies–Applications and Theory.61–67.
- Yinxi Liu Kai Yang Shaoyu Dou and Pan Luo. 2022. Interpretable anomaly detection in variable-length co-evolving rhythmic sequences. In 1st Workshop on Artificial Intelligence for Anomalies and Novelties (AI4AN) .
- Scott M. Lundberg and Su-In Lee. 2017. A unified approach to interpreting model predictions. In Proceedings of the 31st International Conference on Neural Information Processing Systems (Long Beach California USA) (NIPS’17) . Curran Associates Inc. Red Hook NY USA 4768–4777.
- Alireza Makhzani Jonathon Shlens Navdeep Jaitly Ian Goodfellow and Brendan Frey. 2016. Adversarial autoencoders. ICLR Workshop track .
- Guansong Pang, Longbing Cao, and Ling Chen. 2016. Outlier detection in complex categorical data by modelling the feature value couplings. In Proceedings of the IJCAI International Joint Conference on Artificial Intelligence.
- Tomáš Pevnỳ and Martin Kopp. 2014. Explaining anomalies with sapling random forests. In Proceedings of the Information Technologies-Applications and Theory Workshops, Posters, and Tutorials (ITAT’14).
- Daniel M. Roy and Yee Whye Teh. 2008. The mondrian process. In Proceedings of the NIPS, Vol. 21.
- Bernhard Schölkopf Robert C. Williamson Alex Smola John Shawe-Taylor and John Platt. 1999. Support vector method for novelty detection. In Proceedings of the 12th International Conference on Neural Information Processing Systems (Denver CO) (NIPS’99 Vol. 12) . MIT Press Cambridge MA USA 582–588.
- John Sipple. 2020. Interpretable, multidimensional, multimodal anomaly detection with negative sampling for detection of device failure. In Proceedings of the International Conference on Machine Learning. PMLR, 9016–9025.
- Mukund Sundararajan and Amir Najmi. 2020. The many Shapley values for model explanation. In Proceedings of the International Conference on Machine Learning. PMLR, 9269–9278.
- Mukund Sundararajan, Ankur Taly, and Qiqi Yan. 2017. Axiomatic attribution for deep networks. In Proceedings of the International Conference on Machine Learning. PMLR, 3319–3328.
Cited by
Anomaly diagnosis of connected autonomous vehicles: A survey
Fang, Y., Min, H., Wu, X., Wang, W., Zhao, X., Martínez‐Pastor, B., & Teixeira, R. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102223 · Cited by 56
Machine Learning Applications in Manufacturing—Challenges, Trends, and Future Directions
Manta-Costa, A., Araújo, S. O., Peres, R. S., & Barata, J. (2024) · IEEE Open Journal of the Industrial Electronics Society · doi:10.1109/ojies.2024.3431240 · Cited by 66
“As of today, explainable AD has become a regulatory and ethical requirement in its applications [76].”
BackgroundLLM-LADE: Large language model-based log anomaly detection with explanation
Zhang, Z., Li, S., Zhang, L., Ye, J., Hu, C., & Yan, L. (2025) · Knowledge-Based Systems · doi:10.1016/j.knosys.2025.114064 · Cited by 42
Dual GroupGAN: An unsupervised four-competitor (2V2) approach for video anomaly detection
Sun, Z., Wang, P., Wang, Z., & Zhang, M. (2024) · Pattern Recognition · doi:10.1016/j.patcog.2024.110500 · Cited by 39
Graph Neural Networks based Log Anomaly Detection and Explanation
Li, Z., Shi, J., & van Leeuwen, M. (2024) · ACM/IEEE International Conference on Software Engineering: Companion Proceedings (ICSE-Companion) · doi:10.1145/3639478.3643084 · Cited by 43
Self-citation
Metadata
- Authors
- Li, Z., Zhu, Y., & Van Leeuwen, M.
- Affiliations
- Leiden University (NL)
- Leiden Institute of Advanced Computer Science (LIACS), Leiden University, The Netherlands
- Venue
- ACM Transactions on Knowledge Discovery from Data
- Published
- 2023-09-06
- Volume
- 18
- Issue
- 1
- Pages
- 1-54
- ISSN
- 1556-4681 (ISSN-L)
- 1556-472X
- OpenAlex
- W4384407578
- Cited by
- 139
- References
- 203
- Citations by year
- 2026: 47 · 2025: 50 · 2024: 36 · 2023: 6
- FWCI
- 13.07
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- ACM Transactions on Knowledge Discovery from Data (publishedVersion, cc-by-sa)
- arXiv (Cornell University) (submittedVersion)
- Leiden Repository (Leiden University) (submittedVersion)
- arXiv (Cornell University)
- Funders
- Dutch Research Council
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Artificial Intelligence › Anomaly Detection Techniques and Applications
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Keywords
- explainable anomaly detection, anomaly detection, explainability, safety-critical domains
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.
The claim verdicts this report counts are experimental: their accuracy is not yet measured.
A language model drew the claims from the paper on 2026-10-08, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-08 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).
Cited sources are checked with a language model’s help (gpt-6-sol).
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
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.