The NIH Office of Portfolio Analysis has developed a neural network approach to analysis of scientific content using dimensionality reduction (word2vecOPA). This method computationally converts words in scientific texts to numbers and summarizes documents by their semantic content by learning relationships between words from their context. This method is adaptable to specific corpora, including grants and scientific articles. For more information see the publication describing our word2vec approach: Hoppe et al 2019 (https://www.science.org/doi/10.1126/sciadv.aaw7238)
Date
Oct 2022
Source URL
Organization Type
Government