Universe

skweak

Weak supervision for NLP

skweak brings the power of weak supervision to NLP tasks, and in particular sequence labelling and text classification. Instead of annotating documents by hand, skweak allows you to define labelling functions to automatically label your documents, and then aggregate their results using a statistical model that estimates the accuracy and confusions of each labelling function.

Example

import spacy, re from skweak import heuristics, gazetteers, aggregation, utils # LF 1: heuristic to detect occurrences of MONEY entities def money_detector(doc): for tok in doc[1:]: if tok.text[0].isdigit() and tok.nbor(-1).is_currency: yield tok.i-1, tok.i+1, 'MONEY' lf1 = heuristics.FunctionAnnotator('money', money_detector) # LF 2: detection of years with a regex lf2= heuristics.TokenConstraintAnnotator ('years', lambda tok: re.match('(19|20)\d{2}$', tok.text), 'DATE') # LF 3: a gazetteer with a few names NAMES = [('Barack', 'Obama'), ('Donald', 'Trump'), ('Joe', 'Biden')] trie = gazetteers.Trie(NAMES) lf3 = gazetteers.GazetteerAnnotator('presidents', {'PERSON':trie}) # We create a corpus (here with a single text) nlp = spacy.load('en_core_web_sm') doc = nlp('Donald Trump paid $750 in federal income taxes in 2016') # apply the labelling functions doc = lf3(lf2(lf1(doc))) # and aggregate them hmm = aggregation.HMM('hmm', ['PERSON', 'DATE', 'MONEY']) hmm.fit_and_aggregate([doc]) # we can then visualise the final result (in Jupyter) utils.display_entities(doc, 'hmm')

View more
Author info

Pierre Lison

GitHubhttps://github.com/NorskRegnesentral/skweak

Categories pipeline standalone research training

Submit your project

If you have a project that you want the spaCy community to make use of, you can suggest it by submitting a pull request to the spaCy website repository. The Universe database is open-source and collected in a simple JSON file. For more details on the formats and available fields, see the documentation. Looking for inspiration your own spaCy plugin or extension? Check out the project idea label on the issue tracker.

Read the docsJSON source