ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning
Time series anomaly detection in streaming environments faces persistent challenges due to concept drift, which gradually degrades model reliability. In this paper, we propose Anomaly Detection with Drift-Aware Ensemble-based Incremental Learning (ADDAEIL), an unsupervised anomaly detection framewor...
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2025-06-01
|
| Series: | Algorithms |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1999-4893/18/6/359 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1849423973101600768 |
|---|---|
| author | Danlei Li Nirmal-Kumar C. Nair Kevin I-Kai Wang |
| author_facet | Danlei Li Nirmal-Kumar C. Nair Kevin I-Kai Wang |
| author_sort | Danlei Li |
| collection | DOAJ |
| description | Time series anomaly detection in streaming environments faces persistent challenges due to concept drift, which gradually degrades model reliability. In this paper, we propose Anomaly Detection with Drift-Aware Ensemble-based Incremental Learning (ADDAEIL), an unsupervised anomaly detection framework that incrementally adapts to concept drift in non-stationary streaming time series data. ADDAEIL integrates a hybrid drift detection mechanism that combines statistical distribution tests with structural-based performance evaluation of base detectors in Isolation Forest. This design enables unsupervised detection and continuous adaptation to evolving data patterns. Based on the estimated drift intensity, an adaptive update strategy selectively replaces degraded base detectors. This allows the anomaly detection model to incorporate new information while preserving useful historical behavior. Experiments on both real-world and synthetic datasets show that ADDAEIL consistently outperforms existing state-of-the-art methods and maintains robust long-term performance in non-stationary data streams. |
| format | Article |
| id | doaj-art-083d0cf574b5470180ff9e723d5255f6 |
| institution | Kabale University |
| issn | 1999-4893 |
| language | English |
| publishDate | 2025-06-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Algorithms |
| spelling | doaj-art-083d0cf574b5470180ff9e723d5255f62025-08-20T03:30:24ZengMDPI AGAlgorithms1999-48932025-06-0118635910.3390/a18060359ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental LearningDanlei Li0Nirmal-Kumar C. Nair1Kevin I-Kai Wang2Department of Electrical, Computer, and Software Engineering, The University of Auckland, 20 Symonds Street, Auckland CBD, Auckland 1010, New ZealandDepartment of Electrical, Computer, and Software Engineering, The University of Auckland, 20 Symonds Street, Auckland CBD, Auckland 1010, New ZealandDepartment of Electrical, Computer, and Software Engineering, The University of Auckland, 20 Symonds Street, Auckland CBD, Auckland 1010, New ZealandTime series anomaly detection in streaming environments faces persistent challenges due to concept drift, which gradually degrades model reliability. In this paper, we propose Anomaly Detection with Drift-Aware Ensemble-based Incremental Learning (ADDAEIL), an unsupervised anomaly detection framework that incrementally adapts to concept drift in non-stationary streaming time series data. ADDAEIL integrates a hybrid drift detection mechanism that combines statistical distribution tests with structural-based performance evaluation of base detectors in Isolation Forest. This design enables unsupervised detection and continuous adaptation to evolving data patterns. Based on the estimated drift intensity, an adaptive update strategy selectively replaces degraded base detectors. This allows the anomaly detection model to incorporate new information while preserving useful historical behavior. Experiments on both real-world and synthetic datasets show that ADDAEIL consistently outperforms existing state-of-the-art methods and maintains robust long-term performance in non-stationary data streams.https://www.mdpi.com/1999-4893/18/6/359time series anomaly detectionconcept drift adaptationisolation forestincremental learningunsupervised methodedge computing |
| spellingShingle | Danlei Li Nirmal-Kumar C. Nair Kevin I-Kai Wang ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning Algorithms time series anomaly detection concept drift adaptation isolation forest incremental learning unsupervised method edge computing |
| title | ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning |
| title_full | ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning |
| title_fullStr | ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning |
| title_full_unstemmed | ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning |
| title_short | ADDAEIL: Anomaly Detection with Drift-Aware Ensemble-Based Incremental Learning |
| title_sort | addaeil anomaly detection with drift aware ensemble based incremental learning |
| topic | time series anomaly detection concept drift adaptation isolation forest incremental learning unsupervised method edge computing |
| url | https://www.mdpi.com/1999-4893/18/6/359 |
| work_keys_str_mv | AT danleili addaeilanomalydetectionwithdriftawareensemblebasedincrementallearning AT nirmalkumarcnair addaeilanomalydetectionwithdriftawareensemblebasedincrementallearning AT kevinikaiwang addaeilanomalydetectionwithdriftawareensemblebasedincrementallearning |