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

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

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.

Original language	English
Title of host publication	Advances in Knowledge Discovery and Data Mining - 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Proceedings
Editors	Xintao Wu, Myra Spiliopoulou, Can Wang, Vipin Kumar, Longbing Cao, Yanqiu Wu, Zhangkai Wu, Yu Yao
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	95-107
Number of pages	13
ISBN (Print)	9789819681693
DOIs	https://doi.org/10.1007/978-981-96-8170-9_8
Publication status	Published - 1 Jan 2025
Event	29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025 - Sydney, Australia Duration: 10 Jun 2025 → 13 Jun 2025

Publication series

Name	Lecture Notes in Computer Science
Volume	15870 LNCS
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025
Country/Territory	Australia
City	Sydney
Period	10/06/25 → 13/06/25

Keywords

Anomaly detection
Data streams
Ensembles

Access to Document

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

Cite this

Liu, J. J., Cassales, G. W., Liu, F. T., Pfahringer, B., & Bifet, A. (2025). Streaming Isolation Forest. In 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 (pp. 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. editor / Xintao Wu ; Myra Spiliopoulou ; Can Wang ; Vipin Kumar ; Longbing Cao ; Yanqiu Wu ; Zhangkai Wu ; Yu Yao. Springer Science and Business Media Deutschland GmbH, 2025. pp. 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. in 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, pp. 95-107, 29th Pacific-Asia Conference on Knowledge Discovery and Data Mining, PAKDD 2025, Sydney, Australia, 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).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

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. In Wu X, Spiliopoulou M, Wang C, Kumar V, Cao L, Wu Y, Wu Z, Yao Y, editors, 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

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this