Wisdom of the contexts: active ensemble learning for contextual anomaly detection. Calikus, E., Nowaczyk, S., Bouguelia, M., & Dikmen, O. Data Mining and Knowledge Discovery, 36(6):2410–2458, November, 2022. ![Wisdom of the contexts: active ensemble learning for contextual anomaly detection [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex In contextual anomaly detection, an object is only considered anomalous within a specific context. Most existing methods use a single context based on a set of user-specified contextual features. However, identifying the right context can be very challenging in practice, especially in datasets with a large number of attributes. Furthermore, in real-world systems, there might be multiple anomalies that occur in different contexts and, therefore, require a combination of several “useful” contexts to unveil them. In this work, we propose a novel approach, called wisdom of the contexts (WisCon), to effectively detect complex contextual anomalies in situations where the true contextual and behavioral attributes are unknown. Our method constructs an ensemble of multiple contexts, with varying importance scores, based on the assumption that not all useful contexts are equally so. We estimate the importance of each context using an active learning approach with a novel query strategy. Experiments show that WisCon significantly outperforms existing baselines in different categories (i.e., active learning methods, unsupervised contextual and non-contextual anomaly detectors) on 18 datasets. Furthermore, the results support our initial hypothesis that there is no single perfect context that successfully uncovers all kinds of contextual anomalies, and leveraging the “wisdom” of multiple contexts is necessary.
@article{calikus_wisdom_2022,
title = {Wisdom of the contexts: active ensemble learning for contextual anomaly detection},
volume = {36},
issn = {1573-756X},
shorttitle = {Wisdom of the contexts},
url = {https://doi.org/10.1007/s10618-022-00868-7},
doi = {10.1007/s10618-022-00868-7},
abstract = {In contextual anomaly detection, an object is only considered anomalous within a specific context. Most existing methods use a single context based on a set of user-specified contextual features. However, identifying the right context can be very challenging in practice, especially in datasets with a large number of attributes. Furthermore, in real-world systems, there might be multiple anomalies that occur in different contexts and, therefore, require a combination of several “useful” contexts to unveil them. In this work, we propose a novel approach, called wisdom of the contexts (WisCon), to effectively detect complex contextual anomalies in situations where the true contextual and behavioral attributes are unknown. Our method constructs an ensemble of multiple contexts, with varying importance scores, based on the assumption that not all useful contexts are equally so. We estimate the importance of each context using an active learning approach with a novel query strategy. Experiments show that WisCon significantly outperforms existing baselines in different categories (i.e., active learning methods, unsupervised contextual and non-contextual anomaly detectors) on 18 datasets. Furthermore, the results support our initial hypothesis that there is no single perfect context that successfully uncovers all kinds of contextual anomalies, and leveraging the “wisdom” of multiple contexts is necessary.},
language = {en},
number = {6},
urldate = {2023-05-21},
journal = {Data Mining and Knowledge Discovery},
author = {Calikus, Ece and Nowaczyk, Sławomir and Bouguelia, Mohamed-Rafik and Dikmen, Onur},
month = nov,
year = {2022},
keywords = {Active learning, Anomaly detection, Contextual anomaly detection, Ensemble learning},
pages = {2410--2458},
}
Downloads: 0
{"_id":"YSbvxJuPYk669rrcc","bibbaseid":"calikus-nowaczyk-bouguelia-dikmen-wisdomofthecontextsactiveensemblelearningforcontextualanomalydetection-2022","author_short":["Calikus, E.","Nowaczyk, S.","Bouguelia, M.","Dikmen, O."],"bibdata":{"bibtype":"article","type":"article","title":"Wisdom of the contexts: active ensemble learning for contextual anomaly detection","volume":"36","issn":"1573-756X","shorttitle":"Wisdom of the contexts","url":"https://doi.org/10.1007/s10618-022-00868-7","doi":"10.1007/s10618-022-00868-7","abstract":"In contextual anomaly detection, an object is only considered anomalous within a specific context. Most existing methods use a single context based on a set of user-specified contextual features. However, identifying the right context can be very challenging in practice, especially in datasets with a large number of attributes. Furthermore, in real-world systems, there might be multiple anomalies that occur in different contexts and, therefore, require a combination of several “useful” contexts to unveil them. In this work, we propose a novel approach, called wisdom of the contexts (WisCon), to effectively detect complex contextual anomalies in situations where the true contextual and behavioral attributes are unknown. Our method constructs an ensemble of multiple contexts, with varying importance scores, based on the assumption that not all useful contexts are equally so. We estimate the importance of each context using an active learning approach with a novel query strategy. Experiments show that WisCon significantly outperforms existing baselines in different categories (i.e., active learning methods, unsupervised contextual and non-contextual anomaly detectors) on 18 datasets. Furthermore, the results support our initial hypothesis that there is no single perfect context that successfully uncovers all kinds of contextual anomalies, and leveraging the “wisdom” of multiple contexts is necessary.","language":"en","number":"6","urldate":"2023-05-21","journal":"Data Mining and Knowledge Discovery","author":[{"propositions":[],"lastnames":["Calikus"],"firstnames":["Ece"],"suffixes":[]},{"propositions":[],"lastnames":["Nowaczyk"],"firstnames":["Sławomir"],"suffixes":[]},{"propositions":[],"lastnames":["Bouguelia"],"firstnames":["Mohamed-Rafik"],"suffixes":[]},{"propositions":[],"lastnames":["Dikmen"],"firstnames":["Onur"],"suffixes":[]}],"month":"November","year":"2022","keywords":"Active learning, Anomaly detection, Contextual anomaly detection, Ensemble learning","pages":"2410–2458","bibtex":"@article{calikus_wisdom_2022,\n\ttitle = {Wisdom of the contexts: active ensemble learning for contextual anomaly detection},\n\tvolume = {36},\n\tissn = {1573-756X},\n\tshorttitle = {Wisdom of the contexts},\n\turl = {https://doi.org/10.1007/s10618-022-00868-7},\n\tdoi = {10.1007/s10618-022-00868-7},\n\tabstract = {In contextual anomaly detection, an object is only considered anomalous within a specific context. Most existing methods use a single context based on a set of user-specified contextual features. However, identifying the right context can be very challenging in practice, especially in datasets with a large number of attributes. Furthermore, in real-world systems, there might be multiple anomalies that occur in different contexts and, therefore, require a combination of several “useful” contexts to unveil them. In this work, we propose a novel approach, called wisdom of the contexts (WisCon), to effectively detect complex contextual anomalies in situations where the true contextual and behavioral attributes are unknown. Our method constructs an ensemble of multiple contexts, with varying importance scores, based on the assumption that not all useful contexts are equally so. We estimate the importance of each context using an active learning approach with a novel query strategy. Experiments show that WisCon significantly outperforms existing baselines in different categories (i.e., active learning methods, unsupervised contextual and non-contextual anomaly detectors) on 18 datasets. Furthermore, the results support our initial hypothesis that there is no single perfect context that successfully uncovers all kinds of contextual anomalies, and leveraging the “wisdom” of multiple contexts is necessary.},\n\tlanguage = {en},\n\tnumber = {6},\n\turldate = {2023-05-21},\n\tjournal = {Data Mining and Knowledge Discovery},\n\tauthor = {Calikus, Ece and Nowaczyk, Sławomir and Bouguelia, Mohamed-Rafik and Dikmen, Onur},\n\tmonth = nov,\n\tyear = {2022},\n\tkeywords = {Active learning, Anomaly detection, Contextual anomaly detection, Ensemble learning},\n\tpages = {2410--2458},\n}\n\n","author_short":["Calikus, E.","Nowaczyk, S.","Bouguelia, M.","Dikmen, O."],"key":"calikus_wisdom_2022","id":"calikus_wisdom_2022","bibbaseid":"calikus-nowaczyk-bouguelia-dikmen-wisdomofthecontextsactiveensemblelearningforcontextualanomalydetection-2022","role":"author","urls":{"Paper":"https://doi.org/10.1007/s10618-022-00868-7"},"keyword":["Active learning","Anomaly detection","Contextual anomaly detection","Ensemble learning"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://bibbase.org/zotero/mh_lenguyen","dataSources":["iwKepCrWBps7ojhDx"],"keywords":["active learning","anomaly detection","contextual anomaly detection","ensemble learning"],"search_terms":["wisdom","contexts","active","ensemble","learning","contextual","anomaly","detection","calikus","nowaczyk","bouguelia","dikmen"],"title":"Wisdom of the contexts: active ensemble learning for contextual anomaly detection","year":2022}