explosion/spaCy
1
1
Fork 0
mirror of https://github.com/explosion/spaCy.git synced 2024-12-25 17:36:30 +03:00
Code Issues Packages Projects Releases Wiki Activity

Avoid making prepositions get the tag SCONJ

Browse Source
This commit is contained in:
Matthew Honnibal 2019-08-25 21:56:47 +02:00
parent 22250cf6b7
commit 095c63c6b8
1 changed files with 26 additions and 22 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

48
spacy/lang/en/morph_rules.py
View File

@ -3,55 +3,59 @@ from __future__ import unicode_literals
from ...symbols import LEMMA, PRON_LEMMA from ...symbols import LEMMA, PRON_LEMMA
# Several entries here look pretty suspicious. These will get the POS SCONJ
# given the tag IN, when an adpositional reading seems much more likely for
# a lot of these prepositions. I'm not sure what I was running in 04395ffa4
# when I did this? It doesn't seem right.
_subordinating_conjunctions = [ _subordinating_conjunctions = [
"that", "that",
"if", "if",
"as", "as",
"because", "because",
"of", #"of",
"for", #"for",
"before", #"before",
"in", #"in",
"while", "while",
"after", #"after",
"since", "since",
"like", "like",
"with", #"with",
"so", "so",
"to", #"to",
"by", #"by",
"on", #"on",
"about", #"about",
"than", "than",
"whether", "whether",
"although", "although",
"from", #"from",
"though", "though",
"until", #"until",
"unless", "unless",
"once", "once",
"without", #"without",
"at", #"at",
"into", #"into",
"cause", "cause",
"over", #"over",
"upon", "upon",
"till", "till",
"whereas", "whereas",
"beyond", #"beyond",
"whilst", "whilst",
"except", "except",
"despite", "despite",
"wether", "wether",
"then", #"then",
"but", "but",
"becuse", "becuse",
"whie", "whie",
"below", #"below",
"against", #"against",
"it", "it",
"w/out", "w/out",
"toward", #"toward",
"albeit", "albeit",
"save", "save",
"besides", "besides",
@ -63,7 +67,7 @@ _subordinating_conjunctions = [
"out", "out",
"near", "near",
"seince", "seince",
"towards", #"towards",
"tho", "tho",
"sice", "sice",
"will", "will",