wikiclass.features – Revision feature extractors

class wikiclass.features.WikitextAndInfonoise(language)

Constructs a wikitext and information noise feature extractor.

Parameters:language : Language

FEATURES = ['loglength', 'logreferences', 'logpagelinks', 'numimageslength', 'num_citetemplates', 'lognoncitetemplates', 'num_categories', 'infonoisescore', 'hasinfobox', 'lvl2headings', 'lvl3headings']

Order of the columns

extract(text)

Extracts a set of features from text.

Base classes

class wikiclass.features.FeatureExtractor

class wikiclass.features.TextFeatureExtractor