Streaming Isolation Forest

Justin Jia Liu; Guilherme Weigert Cassales; Fei Tony Liu; Bernhard Pfahringer; Albert Bifet

doi:10.1007/978-981-96-8170-9_8

Streaming Isolation Forest

Justin Jia Liu
, Guilherme Weigert Cassales
, Fei Tony Liu
, Bernhard Pfahringer
, Albert Bifet

University of Waikato
Artificial General Intelligence Pty Ltd.
University of New South Wales

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

Résumé

Anomaly detection is crucial to identify unusual patterns in various domains. In particular, continuous and rapid flow creates distinct challenges within streaming data. This paper introduces the Streaming Isolation Forest (SiForest), a novel algorithm that uses isolation principles and reservoir sampling to align the model with current data distributions. SiForest efficiently detects anomalies with minimal computational and memory requirements and dynamically updates its model using a subtree regrowing strategy. Empirical evaluation on twenty-three benchmark datasets demonstrates that SiForest outperforms eight state-of-the-art algorithms in terms of AUC-ROC scores, achieving greater precision and adaptability.

langue originale	Anglais
titre	Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings
rédacteurs en chef	Xintao Wu, Myra Spiliopoulou, Can Wang, Vipin Kumar, Longbing Cao, Yanqiu Wu, Zhangkai Wu, Yu Yao
Editeur	Springer Science and Business Media Deutschland GmbH
Pages	95-107
Nombre de pages	13
ISBN (imprimé)	9789819681693
Les DOIs	https://doi.org/10.1007/978-981-96-8170-9_8
état	Publié - 1 janv. 2025
Evénement	29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025 - Sydney, Australie Durée: 10 juin 2025 → 13 juin 2025

Série de publications

Nom	Lecture Notes in Computer Science
Volume	15870 LNCS
ISSN (imprimé)	0302-9743
ISSN (Electronique)	1611-3349

Une conférence

Une conférence	29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025
Pays/Territoire	Australie
La ville	Sydney
période	10/06/25 → 13/06/25

Accès au document

10.1007/978-981-96-8170-9_8

Autres fichiers et liens

Lien vers la publication dans Scopus

Contient cette citation

Liu, J. J., Cassales, G. W., Liu, F. T., Pfahringer, B., & Bifet, A. (2025). Streaming Isolation Forest. Dans X. Wu, M. Spiliopoulou, C. Wang, V. Kumar, L. Cao, Y. Wu, Z. Wu, & Y. Yao (eds.), Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings (p. 95-107). (Lecture Notes in Computer Science; Vol 15870 LNCS). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-981-96-8170-9_8

Liu, Justin Jia ; Cassales, Guilherme Weigert ; Liu, Fei Tony et al. / Streaming Isolation Forest. Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings. Editeur / Xintao Wu ; Myra Spiliopoulou ; Can Wang ; Vipin Kumar ; Longbing Cao ; Yanqiu Wu ; Zhangkai Wu ; Yu Yao. Springer Science and Business Media Deutschland GmbH, 2025. p. 95-107 (Lecture Notes in Computer Science).

@inproceedings{4f6891eeadfc4573ba1f71d024d0db1d,

title = "Streaming Isolation Forest",

abstract = "Anomaly detection is crucial to identify unusual patterns in various domains. In particular, continuous and rapid flow creates distinct challenges within streaming data. This paper introduces the Streaming Isolation Forest (SiForest), a novel algorithm that uses isolation principles and reservoir sampling to align the model with current data distributions. SiForest efficiently detects anomalies with minimal computational and memory requirements and dynamically updates its model using a subtree regrowing strategy. Empirical evaluation on twenty-three benchmark datasets demonstrates that SiForest outperforms eight state-of-the-art algorithms in terms of AUC-ROC scores, achieving greater precision and adaptability.",

keywords = "Anomaly detection, Data streams, Ensembles",

author = "Liu, \{Justin Jia\} and Cassales, \{Guilherme Weigert\} and Liu, \{Fei Tony\} and Bernhard Pfahringer and Albert Bifet",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.; 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025 ; Conference date: 10-06-2025 Through 13-06-2025",

year = "2025",

month = jan,

day = "1",

doi = "10.1007/978-981-96-8170-9\_8",

language = "English",

isbn = "9789819681693",

series = "Lecture Notes in Computer Science",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "95--107",

editor = "Xintao Wu and Myra Spiliopoulou and Can Wang and Vipin Kumar and Longbing Cao and Yanqiu Wu and Zhangkai Wu and Yu Yao",

booktitle = "Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings",

}

Liu, JJ, Cassales, GW, Liu, FT, Pfahringer, B & Bifet, A 2025, Streaming Isolation Forest. Dans X Wu, M Spiliopoulou, C Wang, V Kumar, L Cao, Y Wu, Z Wu & Y Yao (eds), Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings. Lecture Notes in Computer Science, VOL. 15870 LNCS, Springer Science and Business Media Deutschland GmbH, p. 95-107, 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Sydney, Australie, 10/06/25. https://doi.org/10.1007/978-981-96-8170-9_8

Streaming Isolation Forest. / Liu, Justin Jia; Cassales, Guilherme Weigert; Liu, Fei Tony et al.
Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings. Ed. / Xintao Wu; Myra Spiliopoulou; Can Wang; Vipin Kumar; Longbing Cao; Yanqiu Wu; Zhangkai Wu; Yu Yao. Springer Science and Business Media Deutschland GmbH, 2025. p. 95-107 (Lecture Notes in Computer Science; Vol 15870 LNCS).

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collection › Contribution à une conférence › Revue par des pairs

TY - GEN

T1 - Streaming Isolation Forest

AU - Liu, Justin Jia

AU - Cassales, Guilherme Weigert

AU - Liu, Fei Tony

AU - Pfahringer, Bernhard

AU - Bifet, Albert

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd. 2025.

PY - 2025/1/1

Y1 - 2025/1/1

N2 - Anomaly detection is crucial to identify unusual patterns in various domains. In particular, continuous and rapid flow creates distinct challenges within streaming data. This paper introduces the Streaming Isolation Forest (SiForest), a novel algorithm that uses isolation principles and reservoir sampling to align the model with current data distributions. SiForest efficiently detects anomalies with minimal computational and memory requirements and dynamically updates its model using a subtree regrowing strategy. Empirical evaluation on twenty-three benchmark datasets demonstrates that SiForest outperforms eight state-of-the-art algorithms in terms of AUC-ROC scores, achieving greater precision and adaptability.

AB - Anomaly detection is crucial to identify unusual patterns in various domains. In particular, continuous and rapid flow creates distinct challenges within streaming data. This paper introduces the Streaming Isolation Forest (SiForest), a novel algorithm that uses isolation principles and reservoir sampling to align the model with current data distributions. SiForest efficiently detects anomalies with minimal computational and memory requirements and dynamically updates its model using a subtree regrowing strategy. Empirical evaluation on twenty-three benchmark datasets demonstrates that SiForest outperforms eight state-of-the-art algorithms in terms of AUC-ROC scores, achieving greater precision and adaptability.

KW - Anomaly detection

KW - Data streams

KW - Ensembles

UR - https://www.scopus.com/pages/publications/105009256590

U2 - 10.1007/978-981-96-8170-9_8

DO - 10.1007/978-981-96-8170-9_8

M3 - Conference contribution

AN - SCOPUS:105009256590

SN - 9789819681693

T3 - Lecture Notes in Computer Science

SP - 95

EP - 107

BT - Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings

A2 - Wu, Xintao

A2 - Spiliopoulou, Myra

A2 - Wang, Can

A2 - Kumar, Vipin

A2 - Cao, Longbing

A2 - Wu, Yanqiu

A2 - Wu, Zhangkai

A2 - Yao, Yu

PB - Springer Science and Business Media Deutschland GmbH

T2 - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025

Y2 - 10 June 2025 through 13 June 2025

ER -

Liu JJ, Cassales GW, Liu FT, Pfahringer B, Bifet A. Streaming Isolation Forest. Dans Wu X, Spiliopoulou M, Wang C, Kumar V, Cao L, Wu Y, Wu Z, Yao Y, rédacteurs en chef, Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings. Springer Science and Business Media Deutschland GmbH. 2025. p. 95-107. (Lecture Notes in Computer Science). doi: 10.1007/978-981-96-8170-9_8

Streaming Isolation Forest

Résumé

Série de publications

Une conférence

Accès au document

Autres fichiers et liens

Empreinte digitale

Contient cette citation