An Adaptive Machine Learning Model for Heterogeneous Concept Drift Handling in Streaming Data
Engineering, Technology & Applied Science Research · 04.04.2026 · 16 (2) · 33672-33677
DOI: 10.48084/etasr.17430 ↗Kurz gefasst
Fachartikel in der Zeitschrift Engineering, Technology & Applied Science Research, online erschienen am 04.04.2026. Verfasst von Deepa C. Mulimani und 1 weiteren Autor:innen. Die Arbeit nennt 24 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
The proliferation of streaming data in the present domain intrusion detection systems, financial applications, Internet of Things, and healthcare has transformed the requirements of machine learning applications. Concept drift is a difficult problem in these dynamic environments where the statistical properties of the data change over time, leading to the degradation of machine learning models. This problem is bigger in data streams with multiple classes. Present adaptive learning models are less effective in a variety of concept drift scenarios because they are reactive and have homogeneous drift detection. This article introduces the novel Weighted Adaptive Ensemble Method (WAEM) to overcome these drawbacks. The technique uses a set of different drift detectors and an ensemble of adaptive classifiers to capture the diverse drift events with a zero-detection delay. WAEM proactively responds to the distributional changes through dynamic weight adjustments that are driven by the individual classifier’s performance. Weighted probabilistic fusion and calibrated confidence are used to estimate the final predictions. Experiments on synthetic and real-world datasets show that WAEM performs better in terms of drift adaptation metrics and classification metrics than the three benchmark adaptive methods they were compared with.
Zitieren
APA
Mulimani, D. C., & Patil, P. R. (2026). An Adaptive Machine Learning Model for Heterogeneous Concept Drift Handling in Streaming Data. Engineering, Technology & Applied Science Research, 16(2), 33672-33677. https://doi.org/10.48084/etasr.17430
Vancouver
Mulimani DC, Patil PR. An Adaptive Machine Learning Model for Heterogeneous Concept Drift Handling in Streaming Data. Engineering, Technology & Applied Science Research. 2026;16(2):33672-33677. doi:10.48084/etasr.17430
BibTeX
@article{Mulimani2026,
author = {Mulimani, Deepa C. and Patil, Prakashgoud R.},
title = {An Adaptive Machine Learning Model for Heterogeneous Concept Drift Handling in Streaming Data},
journal = {Engineering, Technology & Applied Science Research},
year = {2026},
volume = {16},
number = {2},
pages = {33672--33677},
doi = {10.48084/etasr.17430}
}
EUVIMED