Articles | Volume 25, issue 5
https://doi.org/10.5194/hess-25-2543-2021
https://doi.org/10.5194/hess-25-2543-2021
Research article
 | 
18 May 2021
Research article |  | 18 May 2021

Resampling and ensemble techniques for improving ANN-based high-flow forecast accuracy

Everett Snieder, Karen Abogadil, and Usman T. Khan

Viewed

Total article views: 2,419 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
1,710 673 36 2,419 29 27
  • HTML: 1,710
  • PDF: 673
  • XML: 36
  • Total: 2,419
  • BibTeX: 29
  • EndNote: 27
Views and downloads (calculated since 09 Oct 2020)
Cumulative views and downloads (calculated since 09 Oct 2020)

Viewed (geographical distribution)

Total article views: 2,419 (including HTML, PDF, and XML) Thereof 2,238 with geography defined and 181 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 

Cited

Latest update: 21 Apr 2024
Download
Short summary
Flow distributions are highly skewed, resulting in low prediction accuracy of high flows when using artificial neural networks for flood forecasting. We investigate the use of resampling and ensemble techniques to address the problem of skewed datasets to improve high flow prediction. The methods are implemented both independently and in combined, hybrid techniques. This research presents the first analysis of the effects of combining these methods on high flow prediction accuracy.