Technical Report - Detecting Federally Funded Research and Development Trends Using Machine Learning and Information Retrieval Methods

Federal RePORTER, a recently retired federally funded research and development (R&D) grant database, contained a vast amount of information on federally funded R&D and was utilized by researchers, citizens, and policymakers alike to uncover insights. In this report, we provide a classification of research topics contained within Federal RePORTER project abstracts, as well as trends in these topics over time, using natural language processing (NLP) and machine learning techniques. In collaboration with the National Center for Science and Engineering Statistics (NCSES), we examined how the topics and their trends change as a result of the number of topics produced by the model. In addition, we utilized information retrieval techniques to find theme-related topics and their trends over time. This is realized through two case-studies, the first using the theme of pandemics and the second using the theme of coronavirus.

Linehan K, Oh E, Thurston J, Siwe GL, Garrett M, Keller S, Shipp S, Kindlon A, Jankowski, J. Technical Report - Detecting Federally Funded Research and Development Trends Using Machine Learning and Information Retrieval Methods, Technical Report. BI-2022-1531. Proceedings of the Biocomplexity Institute. University of Virginia; 2022 May. DOI: https://doi.org/10.18130/4c3g-k017.