.png){kind=icon}
.png){kind=icon}
Return
Tree-based vs. deep learning in time series: A prescriptive framework for performance-driven model selection
.png){kind=icon}
OA
AI
DOI：10.1016/j.asoc.2026.114983.png)
.png){kind=icon}
.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
• Propose a novel prescriptive framework for time series forecasting model selection. • Present a large-scale benchmark with statistical significance testing across 50,000 time series. • Challenge the notion of a universal best model by linking performance to data characteristics and computational efficiency. • Automate model selection with interpretable rules, accounting for accuracy, efficiency, and sustainability.
Keywords:
Time series forecasting
Prescriptive analytics
Model selection
Interpretable artificial intelligence
Tree-based models
Deep learning
Meta-learning
AI Summary
Key information extracted from the uploaded paper, including a brief overview, abstract, background, key highlights, visual analysis, and future outlook.