I have been using deeppavlov's named entity recognition model, however, it returns data in this format: [[[tokens], [ner_tags]]]
Example:
Raw text- John Doe at Burger King on Thursday
Return:
[[['john', 'doe', 'at', 'burger', 'king', 'on', 'thursday'],
['B-PERSON, 'I-PERSON', 'O', B-ORG, I-ORG, 'O', 'B-DATE]]]
Desired:
[['john doe', 'PERSON'], ['burger king', ORG], [thursday, DATE]]
The 'B-' prefix indicates the beginning of an entity, while 'I-' indicates the 'inside' of the entity. How do I manipulate the lists to provide the desired output
You could use the zip
method.
rs = [[['john', 'doe', 'at', 'burger', 'king', 'on', 'thursday'],
['B-PERSON, 'I-PERSON', 'O', B-ORG, I-ORG, 'O', 'B-DATE]]]
words, kinds = rs[0]
classes = [[word, kind] for word, kind in zip(words, kinds) if kind != 'O']
Use itertools.groupby
:
from itertools import groupby
res = []
for k, g in groupby(zip(*result[0]), key=lambda x:x[1].split('-')[-1]):
if k != 'O':
res.append([' '.join(x[0] for x in g), k])
res
Output:
[['john doe', 'PERSON'], ['burger king', 'ORG'], ['thursday', 'DATE']]
You can make this one-liner:
[[' '.join(x[0] for x in g), k] for k, g in groupby(zip(*result[0]), key=lambda x:x[1].split('-')[-1]) if k != 'O']
The technical post webpages of this site follow the CC BY-SA 4.0 protocol. If you need to reprint, please indicate the site URL or the original address.Any question please contact:yoyou2525@163.com.