nasa-petal / bio-strategy-extractor

The Unlicense
4 stars 1 forks source link

Compile a dataset for generating summaries #13

Open bruffridge opened 2 years ago

bruffridge commented 2 years ago

Get a dataset of biology papers that includes: title, abstract, and DOI. SCOPUS and the GRC librarians may be helpful for this. May want to filter out the biomed category if possible. These papers will be run through a binary classifier that determines if the paper contains a description of a biological strategy, then on to a model that extracts and summarizes the strategy.