consolidates and restructures information extraction functionality under a single textacy.extract subpackage, including the contents of the extract.py module, plus the ke subpackage and text_utils.py module
Functions previously accessible from extract.py have the same signature (e.g. textacy.extract.words()), but others have changed in a way that is more consistent and semantically meaningful (e.g. from textacy import ke; ke.textrank() => from textacy import extract; extract.keyterms.textrank() or from textacy import text_utils; text_utils.keywords_in_context() => from textacy import extract; extract.keywords_in_context()
improves the generality and quality of extracted "triples", and changes the structure of returned objects accordingly
Previously, only contiguous tokens were returned as spans for each element, such as the "verb" in an SVO, but this was overly restrictive: A sentence like "I did not really like the movie." would produce ("I", "like", "movie") which loses a critical aspect of the meaning. The new method gives (["I"], ["did", "not", "like"], ["movie"]), which is more complex but necessarily so.
For convenience, triple results are all named tuples, so constituent elements may be accessed by name or index (e.g. svo.subject == svo[0])
Previously, direct quotations were only extracted for English-language documents; now, they're available for Spanish and French as well.
Changes behavior of extract.keywords_in_context() to always yield results, with optional padding of contexts; previously, users could choose to print results to stdout or yield results, but padding was non-configurable and applied only when printing, which was weird from a usability point of view
Adds new extraction functions
extract.regex_matches(): For matching regex patterns in a document's text that cross over spaCy token boundaries, which rules out using their built-in Matcher class
extract.acronyms(): For extracting acronym-like tokens, without looking around for related definitions.
Removes old and deprecated extract.pos_regex_matches() function, for which spaCy's Matcher class is an appropriate and more powerful tool, easily accessible via extract.token_matches()
Improves documentation, tests, and type annotations throughout
Motivation and Context
How Has This Been Tested?
Screenshots (if appropriate):
Types of changes
[x] Bug fix (non-breaking change which fixes an issue)
[x] New feature (non-breaking change which adds functionality)
[x] Breaking change (fix or feature that would cause existing functionality to change)
Checklist:
[x] My code follows the code style of this project.
[x] My change requires a change to the documentation, and I have updated it accordingly.
Description
textacy.extractsubpackage, including the contents of theextract.pymodule, plus thekesubpackage andtext_utils.pymoduleextract.pyhave the same signature (e.g.textacy.extract.words()), but others have changed in a way that is more consistent and semantically meaningful (e.g.from textacy import ke; ke.textrank()=>from textacy import extract; extract.keyterms.textrank()orfrom textacy import text_utils; text_utils.keywords_in_context()=>from textacy import extract; extract.keywords_in_context()("I", "like", "movie")which loses a critical aspect of the meaning. The new method gives(["I"], ["did", "not", "like"], ["movie"]), which is more complex but necessarily so.svo.subject==svo[0])extract.keywords_in_context()to always yield results, with optional padding of contexts; previously, users could choose to print results tostdoutor yield results, but padding was non-configurable and applied only when printing, which was weird from a usability point of viewextract.regex_matches(): For matching regex patterns in a document's text that cross over spaCy token boundaries, which rules out using their built-inMatcherclassextract.acronyms(): For extracting acronym-like tokens, without looking around for related definitions.extract.pos_regex_matches()function, for which spaCy'sMatcherclass is an appropriate and more powerful tool, easily accessible viaextract.token_matches()Motivation and Context
How Has This Been Tested?
Screenshots (if appropriate):
Types of changes
Checklist: