Ensemble aggregation methods for relocating models of rare events
| Autor: | Greg Timms, Ashfaqur Rahman, Alison R. Turnbull, Claire D'Este |
|---|---|
| Rok vydání: | 2014 |
| Předmět: |
Computer science
business.industry Machine learning computer.software_genre Ensemble learning Class (biology) Aggregation methods Artificial Intelligence Control and Systems Engineering Rare events One-class classification Artificial intelligence Data mining Electrical and Electronic Engineering business computer |
| Zdroj: | Engineering Applications of Artificial Intelligence. 34:58-65 |
| ISSN: | 0952-1976 |
| DOI: | 10.1016/j.engappai.2014.05.007 |
| Popis: | Spatially distributed regions may have different influences that affect the underlying physical processes and make it inappropriate to directly relocate learned models. We may also be aiming to detect rare events for which we have examples in some regions, but not others. Three novel voting methods are presented for combining classifiers trained on regions with available examples for predicting rare events in new regions; specifically the closure of shellfish farms. The ensemble methods introduced are consistently more accurate at predicting closures. Approximately 63% of locations were successfully learned with Class Balance aggregation compared with 37% for the Expert guidelines, and 0% for One Class Classification. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect