Semantic maps and metrics for science using deep transformer encoders

« The growing deluge of scientific publications demands text analysis tools that can help scientists and policy-makers navigate, forecast and beneficially guide scientific research. Recent advances in natural language understanding driven by deep transformer networks offer new possibilities for mapping science. (…)
Here we report a procedure for encoding scientific documents with these tools, measuring their improvement over static word embeddings in a nearest-neighbor retrieval task.(…) »

source >, Brendan Chambers, James Evans, 13 avril 2021, arXiv:2104.05928