This repository accompanies the publication "Scientific literature on carbon dioxide removal revealed as much larger through AI-enhanced systematic mapping", available here.
The processed data can be explored and downloaded from climateliterature.org.
-
Training Models
The training folder contains the code used to train and test the classification models that categorize the literature in the systematic map. -
Model Evaluation
The check_models folder includes scripts to evaluate model performance, generate evaluation metrics, and calculate confidence intervals. -
Figure Generation
The analysis_plots folder provides the code used to create all figures published in the paper.
All code is written in Python and organized for reproducibility. Detailed instructions for training the models can be found in the README file within the respective subfolder.