Virtual Research Assistant (VRA)

The ongoing projects of Virtual Research Assistant (VRA), which is a comprehensive research recommendation system providing recommendations regarding datasets, publications, grants and collaborators for scholars of interest in the population health domain. Developed by researchers from Department of Biostats and Data Science, UTSPH.

1. Main contributions:

2. A. System snapshot:

2. B. Service platform:

File organization

This repo is organized by different components and projects folders:

• dataset_rec: dataset recommendation project, led by Ginny
• grants_rec: grant recommendation project, led by Ginny
• collaborator_rec: collaborator recommendation project, led by Ginny
• conference_rec: conference recommendation project, led by Rachit
• uncertainty_rec: uncertainty quantification incorporated recommender project, led by Ginny
• preparation: initial data processing for the aforementioned project, led by Braja

Getting Started

examples

See each component for detailed usage of codes

Built With

• torch: one of the most commonly used deep learning library
• Pytorch Geometric: pytorch built for graphs data
• Transformers: pytorch library of transformers
• Pandas, numpy, skearn, and other common machine learning packages, see requirements.txt for details

Authors

See also the list of contributors who participated in this project.

License

This project is licensed under the MIT License

Acknowledgments

• Hat tip to TGN repo, the RS repo and another RS repo for providing papers and codes for us to start with initially