Implementation Code: A Supervised Machine Learning Approach for Assessing Grant Peer Review Reports

The implementation code is an output of the study “A Supervised Machine Learning Approach for Assessing Grant Peer Review Reports”, which combines qualitative human annotation and transformer-based machine-learning approaches to assess characteristics of grant proposal peer review reports submitted to the Swiss National Science Foundation (SNSF). The codes implement the classification analyses based on pre-trained models which are fine-tuned or prompted for classification of the set of 12 categories that reflect the content of grant peer review reports.