.png){kind=icon}
.png){kind=icon}
Return
Improving monsoon forecasting based on feature selection and explainable artificial intelligence
.png){kind=icon}
PRE
AI
DOI：10.1016/j.asoc.2025.114053.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• A novel methodology for forecasting extreme Western North Pacific Summer Monsoon events one month ahead, integrating machine learning-based feature selection techniques with explainable artificial intelligence. • The machine learning models trained using automatically selected features demonstrate superior predictive accuracy compared to those relying on expert selected features. • The explainability analysis successfully identifies crucial climate indices influencing monsoon dynamics, enhancing the reliability of the models, fostering transparency and supporting climate science decision-making. • The results uncover key climate indices and time lags that drive monsoon variability.
Journal
IF:
6.6
Papers:
1.4W
Citations:
4.8W
Organization
No organization information available