def text_pos(sentence):
word = extract.segment(sentence)
pos = extract.pos(word)
result = []
for i, item in enumerate(zip(word, pos)):
result.append({'id': i, 'word': item[0], 'pos': item[1]})
return result
def text_ner(sentence):
word = extract.segment(sentence)
pos = extract.pos(word)
ner = extract.ner(word, pos)
result = []
for i, item in enumerate(zip(word, pos, ner)):
result.append({
'id': i,
'word': item[0],
'pos': item[1],
'ne': item[2]
})
return result
def text_srl(sentence):
word = extract.segment(sentence)
pos = extract.pos(word)
_, arcs = extract.parse(word, pos)
roles = extract.srl(word, pos, arcs)
seg = text_seg(sentence)
result = []
for item in roles:
result.append({
'id': item[0],
'type': item[1],
'beg': item[2],
'end': item[3]
})
return {'seg': seg, 'role': result}
def text_parse(sentence):
word = extract.segment(sentence)
pos = extract.pos(word)
parsers, _ = extract.parse(word, pos)
result = []
for i, item in enumerate(zip(word, pos, parsers)):
result.append({
'id': i,
'word': item[0],
'pos': item[1],
'parent': int(item[2][0]) - 1,
'relate': item[2][1]
})
return result
def text_seg(sentence):
word = list(extract.segment(sentence))
result = []
for i, item in enumerate(word):
result.append({'id': i, 'word': item})
return result