Forked from ksopyla/polish_sentence_nltk_tokenizer.py
Last active
May 8, 2020 18:02
-
-
Save Lysander6/d17ca0aa5ecb80b049334e03c467e572 to your computer and use it in GitHub Desktop.
A curated list of Polish abbreviations for NLTK sentence tokenizer based on Wikipedia text
This file contains 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
import nltk | |
# interactive download | |
# nltk.download() | |
nltk.download('punkt') | |
extra_abbreviations = ['ps', 'inc', 'Corp', 'Ltd', 'Co', 'pkt', 'Dz.Ap', 'Jr', 'jr', 'sp', 'Sp', 'poj', 'pseud', 'krypt', 'sygn', 'Dz.U', 'ws', 'itd', 'np', 'sanskryt', 'nr', 'gł', 'Takht', 'tzw', 't.zw', 'ewan', 'tyt', 'oryg', 't.j', 'vs', 'l.mn', 'l.poj' ] | |
position_abbrev = ['Ks', 'Abp', 'abp','bp','dr', 'kard', 'mgr', 'prof', 'zwycz', 'hab', 'arch', 'arch.kraj', 'B.Sc', 'Ph.D', 'lek', 'med', 'n.med', 'bł', 'św', 'hr', 'dziek' ] | |
quantity_abbrev = [ 'mln', 'obr./min','km/godz', 'godz', 'egz', 'ha', 'j.m', 'cal', 'obj', 'alk', 'wag' ] # not added: tys. | |
actions_abbrev = ['tłum','tlum','zob','wym', 'pot', 'ww', 'ogł', 'wyd', 'min', 'm.i', 'm.in', 'in', 'im','muz','tj', 'dot', 'wsp', 'właść', 'właśc', 'przedr', 'czyt', 'proj', 'dosł', 'hist', 'daw', 'zwł', 'zaw' ] | |
place_abbrev = ['Śl', 'płd', 'geogr'] | |
lang_abbrev = ['jęz','fr','franc', 'ukr', 'ang', 'gr', 'hebr', 'czes', 'pol', 'niem', 'arab', 'egip', 'hiszp', 'jap', 'chin', 'kor', 'tyb', 'wiet', 'sum'] | |
military_abbrev = ['kpt', 'kpr', 'obs', 'pil', 'mjr','płk', 'dypl', 'pp', 'gw', 'dyw', 'bryg', 'ppłk', 'mar', 'marsz', 'rez', 'ppor', 'DPanc', 'BPanc', 'DKaw', 'p.uł', 'nadkom'] | |
extra_abbreviations= extra_abbreviations + position_abbrev + quantity_abbrev + place_abbrev + actions_abbrev + place_abbrev + lang_abbrev+military_abbrev | |
sentence_tokenizer = nltk.data.load('tokenizers/punkt/polish.pickle') | |
sentence_tokenizer._params.abbrev_types.update(extra_abbreviations) | |
text = '.....' | |
sentences = sentence_tokenizer.tokenize(text) |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment