#!usr/bin/env python3
# coding: utf-8
"""Automatic annotation of potential relations with semantic label."""
import sys
sys.path.append('../../../../../tools/data-api/derinet-python/')
import derinet_api
# load DeriNet
derinet = derinet_api.DeriNet(sys.argv[2])
def searchLexeme(lem, p=None):
"""Search lemma in DeriNet. Return None if lemma is not in DeriNet."""
def divideWord(word):
"""Return lemma and pos of word in annotated data (sep=ALT+0150)."""
word = word.split('–')
lemma = word[0]
pos = None
if len(word) > 1:
if word[1] != 'None':
pos = word[1]
return lemma, pos
lem, p = divideWord(lem)
candidates = derinet.search_lexemes(lem, pos=p)
if len(candidates) == 0: # not in
return None
#!/usr/bin/env python3
# coding: utf-8
import sys
sys.path.append('../../../../../../tools/data-api/derinet-python/')
import derinet_api
der = derinet_api.DeriNet('../../../../../releases/cs/derinet-1-5-1.tsv')
def checkAmbig(word):
word = word.split('–')
lem = word[0]
p = None
if len(word) > 1:
if word[1] != 'None':
p = word[1]
m = None
if len(word) > 2:
m = word[2]
return der.search_lexemes(lemma=lem, pos=p, morph=m)
for i in range(1, len(sys.argv)):
print(5 * '-', sys.argv[i], 5 * '-')
lnum = 0
def add_verb_deriv(data_folder, suppress_warnings=True, almost_silent=True):
"""Verb derivation dataset."""
for fname in sorted(
filter(
lambda x: x.endswith('_final_populated.tsv') and not x.
startswith('aspect'), os.listdir(data_folder))):
print('\nAdding {}'.format(fname))
with open(os.path.join(data_folder, fname), 'r',
encoding='utf-8') as import_file:
for line in import_file:
line_parts = line.strip().split('\t')
child = get_info(line_parts[2].strip('"'), 'V')
parent = get_info(line_parts[5].strip('"'), 'V')
if not almost_silent:
print('Setting {} <- {}'.format(child, parent))
check_and_add(child,
parent,
suppress_warnings=suppress_warnings)
if __name__ == "__main__":
derinet = derinet_api.DeriNet(original_derinet_fname)
add_changed_edge_probabilities(changed_edge_fname)
correct_changed_edge_probabilities(dvojiti_rodice_fname)
add_jk((jk_fname_01, jk_fname_02))
#add_verb_deriv(verb_data_folder)
derinet.save(fname=new_derinet_fname, sort=True)
#!usr/bin/env python3
# coding: utf-8
"""Lists unmotivated lemmas."""
import sys
sys.path.append('../../../../../../tools/data-api/derinet-python/')
import derinet_api
derinet = derinet_api.DeriNet('../../../../../releases/cs/derinet-1-7.tsv')
def divideWord(word):
"""Return lemma and pos of word in annotated data.
Used for ambiguous words. _ is separator.
"""
word = word.split('_')
lemma = word[0]
pos = None
if len(word) > 1:
if word[1] != 'None':
pos = word[1]
return lemma, pos
def searchLexeme(lem, p=None):
"""Search lemma in DeriNet. Raise warnings for not beeing inside the
DeriNet and for homonymous lemma.
"""