This repo contains NLP modeling code outlined in the following manuscript:
Lea, I.A., Wikoff, D. Borghoff, S., Fitch, S., Chappell, G., Urban, J.D., Perry, C., Choksi, N., Britt, J., Heintz, M., Klaren, W., Chew, R., Edwards, S., Bever, R.J., Hamernik, K., Kirk, A.B., Lynn, S.G., Markey, K.J. (2024) "Development of a systematic evidence mapping workflow and case study application of the workflow to interrogate thyroid hormone network information".
In particular, the notebooks directory contains three relevant files:
- 01_Create_Embeddings.ipynb
- Creates text embeddings from article titles and abstracts using the SPECTER model.
- 02_Prediction_Report.ipynb
- Creates text classification models for the primary fields used in the study.
- 03_Article_Similiarity.ipynb
- Demonstrates article similarity approach used to support article selection.
Tags: systematic evidence mapping, machine learning, NLP, text classification, semantic similarity
To run notebooks, first install the required python packages:
pip install -r requirements.txt
Then use Jupyter to host the notebooks in the browswer:
juypter notebooks
The United States Environmental Protection Agency (EPA) GitHub project code is provided on an "as is" basis and the user assumes responsibility for its use. EPA has relinquished control of the information and no longer has responsibility to protect the integrity, confidentiality, or availability of the information. Any reference to specific commercial products, processes, or services by service mark, trademark, manufacturer, or otherwise, does not constitute or imply their endorsement, recommendation or favoring by EPA. The EPA seal and logo shall not be used in any manner to imply endorsement of any commercial product or activity by EPA or the United States Government.
For questions, please reach out to Scott Lynn at Lynn.Scott@epa.gov