Adaptive Ensemble Machine Learning Models for Predictive Analytics in Dynamic and Uncertain Data Environments
DOI:
https://doi.org/10.66592/kmp782raKeywords:
adaptive ensemble learning; concept drift; streaming analytics; uncertainty estimation; online learning; predictive modeling; dynamic data.Abstract
Predictive systems deployed in evolving environments face concept drift, featuredistribution shifts, incomplete observations, and changing costs of incorrect decisions. This paper proposes an uncertainty-aware adaptive ensemble framework combining heterogeneous online learners, drift monitoring, reliability-weighted aggregation, and selective model updates. The method integrates exponentially discounted predictive loss, uncertainty calibration, a Page–Hinkley-style drift detector, and a bounded modelreplacement policy. A reproducible experimental protocol is specified for streaming classification under gradual, abrupt, and recurring drift. Numerical results and charts are presented as explicitly labeled illustrative scenarios, not as measurements from an executed benchmark; they demonstrate the expected reporting format and testable hypotheses. The framework offers a practical blueprint for assessing predictive accuracy, calibration, recovery time, computational overhead, and operational stability. Its principal contribution is a coherent evaluation design that distinguishes performance under stable data from resilience during distributional change.
Downloads
Published
Issue
Section
License
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
License Terms
Journal of Emerging Intelligence and Engineering Technologies is an open-access journal. Unless otherwise stated, all articles published in the journal are licensed under the Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0).
Full license text: https://creativecommons.org/licenses/by-nc/4.0/
Under this license, you are free to:
- Share: copy and redistribute the material in any medium or format.
- Adapt: remix, transform, and build upon the material.
The licensor cannot revoke these freedoms as long as you follow the license terms.
Under the following terms:
- Attribution: You must give appropriate credit to the original author(s) and the source. Credit should include the article title, author names, journal name, volume, issue, year, and DOI where available. You must also provide a link to the license and indicate whether changes were made. You may do so in any reasonable manner, but not in any way that suggests the licensor or the journal endorses you or your use.
- NonCommercial: You may not use the material for commercial purposes. Commercial purposes are those primarily intended for, or directed toward, commercial advantage or monetary compensation.
- No additional restrictions: You may not apply legal terms or technological measures that legally restrict others from doing anything the license permits.
Notices
- You do not have to comply with the license for parts of the material that are in the public domain, or where your use is permitted by an applicable exception or limitation, such as fair use or fair dealing.
- No warranties are given. The license may not give you all the permissions you need for your intended use. For example, other rights such as publicity, privacy, or moral rights may limit how you use the material.
- Third-party content in an article, such as figures, tables, or images reproduced from other sources, may be under different terms. It is covered by this license only if it is stated in the credit line.
Copyright
Authors keep the copyright of their work. By publishing, they grant the Journal of Emerging Intelligence and Engineering Technologies the right of first publication under the CC BY-NC 4.0 license.
Commercial Use
Anyone who wants to use published material for commercial purposes must get written permission from the copyright holder(s). Requests can be sent to the journal's editorial office at [editorial email address].
