Ensemble learning-based online sequential pre-interference extreme learning for concept drifting and class imbalanced data streams
Yinjie Huang, Hui Wen, Qunhua Tang, Haitao Liu · 2026
Read Ensemble learning-based online sequential pre-interference extreme learning for concept drifting and class imbalanced data streams by Yinjie Huang, Hui Wen, Qunhua Tang, Haitao Liu (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/plos-pone-0353728.txt.
Description
Yinjie Huang and three colleagues published this 2026 machine learning paper on data that arrives as an endless stream. Such streams shift over time, and rare classes get swamped by common ones, so a fixed model quickly goes stale. Their design adds a reshaping layer to an online learning machine, plus a forgetting factor that lets old patterns fade.
A drift alarm based on balanced accuracy watches for change, and several of these learners vote together as an ensemble. Across nine artificial streams and two real ones, the system handled imbalance and shifting patterns better than rivals. Banks, sensors, and hospitals all depend on stream models that stay honest.
Catalog record
| CREATOR | Yinjie Huang, Hui Wen, Qunhua Tang, Haitao Liu |
|---|---|
| YEAR | 2026 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 71316 |
| LICENSE | CC_BY |
| RIGHTS | CC BY |
| SOURCE | PLOS |