def _json_to_instance(self, json_dict: JsonDict) -> Instance:
"""
Expects JSON that looks like ``{"sentence": "..."}``.
"""
if 'text_idx' in json_dict:
x = (json_dict['text_idx'], json_dict['comment_idx']
) # , json_dict['comment_idx'])
return self._dataset_reader.text_to_instance(*x) # type: ignore
return self._dataset_reader.text_to_instance(*json_dict.values())
def align_entities(extracted: List[str],
literals: JsonDict,
stemmer: NltkPorterStemmer) -> List[str]:
"""
Use stemming to attempt alignment between extracted world and given world literals.
If more words align to one world vs the other, it's considered aligned.
"""
literal_keys = list(literals.keys())
literal_values = list(literals.values())
overlaps = [get_stem_overlaps(extract, literal_values, stemmer) for extract in extracted]
worlds = []
for overlap in overlaps:
if overlap[0] > overlap[1]:
worlds.append(literal_keys[0])
elif overlap[0] < overlap[1]:
worlds.append(literal_keys[1])
else:
worlds.append(None)
return worlds
