KGML-ag: a modeling framework of knowledge-guided machine learning to simulate agroecosystems: a case study of estimating N2O emission using data from mesocosm experiments

Keyphrases

• Modeling Framework 100%
• N2O Emission 100%
• Mesocosm Experiment 100%
• Machine Learning Models 100%
• Agroecosystem 100%
• Knowledge-guided Machine Learning 100%
• N2O Flux 36%
• Process-based Model 27%
• Data Demand 18%
• Ecosys 18%
• Cropland 9%
• Agriculture 9%
• Denitrification 9%
• Biogeochemistry 9%
• Machine Learning 9%
• Geoscience 9%
• Nitrification 9%
• Model Learning 9%
• Complex Interactions 9%
• Hierarchical Structure 9%
• Human Activities 9%
• Black Box 9%
• Recent Advances 9%
• Massive Data 9%
• Microbial Processes 9%
• Soil-plant 9%
• Synthetic Data 9%
• Challenging Tasks 9%
• Temporal Dynamics 9%
• Multi-task Learning 9%
• Model Parameter Uncertainty 9%
• Failure Data 9%
• Emission Peak 9%
• Optimized Model 9%
• Plant Activity 9%
• Global Greenhouse Gas 9%
• Greenhouse Gas Budget 9%
• Agroecosystem Model 9%
• Soil Activity 9%
• Chemical Domain 9%
• Simultaneous Learning 9%
• Gated Recurrent Unit 9%
• Interpretable Machine Learning 9%
• Interpretable Prediction 9%
• Model Evaluation 9%
• Nitrous Oxide (N2O) Emissions 9%

Earth and Planetary Sciences

• Nitrous Oxide 100%
• Agricultural Ecosystem 100%
• Machine Learning 100%
• Mesocosm 100%
• Machine Learning Model 100%
• Farmland 9%
• Time Series 9%
• Gated Recurrent Unit 9%
• Greenhouse Gas 9%
• Earth Science 9%
• Biogeochemistry 9%
• Denitrification 9%
• Nitrification 9%
• Multitask Learning 9%
• Interpretable Machine Learning 9%
• Interpretability 9%

Immunology and Microbiology

• Denitrification 100%
• Biogeochemistry 100%
• Nitrification 100%