Carlos-Francisco Méndez-Cruz / nlp-preprocessing-pipeline

Go to a project
Toggle navigation Toggle navigation pinning
Carlos-Francisco Méndez-Cruz
e425a3be 1 parent 339bca7c

New terminological tagging for CRFs

Inline Side-by-side
Showing 1 changed file with 11 additions and 9 deletions
......@@ -136,17 +136,19 @@ if __name__ == "__main__":
if len(word) > 1:
for termTag in hashTerms:
if word in hashTerms[termTag]:
wordOrig = word.replace('-', ' ')
#print("Word: {}".format(word))
if wordOrig in hashTermsOrig[termTag]:
print("WordOrig: {}".format(wordOrig))
line = ''
for w, l in zip(word.split('-'), lemma.split('-')):
line += w + '\t' + listLine1[1] + '\t' + l + ' ' + termTag + ' TermTag' + '\n'
line.rstrip('\n')
if word.find('-') > -1:
wordOrig = word.replace('-', ' ')
#print("Word: {}".format(word))
if wordOrig in hashTermsOrig[termTag]:
print("WordOrig: {}".format(wordOrig))
line = ''
for w, l in zip(word.split('-'), lemma.split('-')):
line += w + '\t' + listLine1[1] + '\t' + l + ' ' + termTag + ' TermTag' + '\n'
line.rstrip('\n')
else:
line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + termTag + ' TermTag'
else:
line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + termTag + ' TermTag'
#line = listLine1[0] + '\t' + termTag + '\t' + listLine2[0] + ' ' + termTag + ' TermTag'
else:
line = listLine1[0] + '\t' + listLine1[1] + '\t' + listLine2[0] + ' ' + 'O' + ' TermTag'
# line = listLine1[0] + '\t' + termTag + '\t' + listLine2[0] + ' ' + termTag + ' TermTag'
......