Repository navigation
Generate training data: match an author's name in whichever spelling the page sets it - #789
Merged
Merged
Conversation
A line of ORCIDs sets `Ana B. Rivera` where the deposit declares `Ana Beatriz Rivera`, the same name with the middle one cut to an initial. A repeat is accepted only where every character of the declared name is there, so names set that way were background while the identifiers beside them each carried a label. GROBID gives that spelling a byline of its own. An author declared with more than one given name is now offered in the shortened spelling as well, keeping the first given name in full, and is looked for under it after the declared one. Over what the name itself has not already taken, because the two spellings share most of their characters and the shortened one would otherwise be found inside the byline the declared one holds. The other three corpora are byte-identical.
The fixtures and the prose explaining them quoted authors, titles and references lifted from the corpora, which names real people in a public repository to make a point that invented names make just as well. Every shape each test turns on is unchanged: the initials separated by a dot, the surname the gap fill bridges, the reference split across a page break.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
part of https://github.com/eLifePathways/ScienceBeam2.0/issues/205
A page may set an author's name with the middle names cut to initials,
Ana B. Riverawhere the deposit declaresAna Beatriz Rivera. A repeat isaccepted only where every character of the declared name is there, so in the
generated header data a name set that way stayed background while the
identifier beside it carried a label. The data knew the ORCID belonged to
someone and could not say who, which is the shape a reader asked about.
An author declared with more than one given name is now looked for under the
shortened spelling as well as the declared one, and GROBID gives that
spelling a byline of its own. It costs nothing on a page that sets the names
out in full, and shortening the middle names is how a page fits a long author
list beside anything else, so what it is worth grows with the corpus.