Reconocimiento de entidades y extracción de relaciones

  • 30 min | Última modificación: Diciembre 9, 2020

http://www.nltk.org/book/

Text Analytics with Python

[1]:

import nltk

nltk.download('maxent_ne_chunker')
nltk.download('words')
nltk.download('ieer')

[nltk_data] Downloading package maxent_ne_chunker to
[nltk_data]     /root/nltk_data...
[nltk_data]   Package maxent_ne_chunker is already up-to-date!
[nltk_data] Downloading package words to /root/nltk_data...
[nltk_data]   Package words is already up-to-date!
[nltk_data] Downloading package ieer to /root/nltk_data...
[nltk_data]   Package ieer is already up-to-date!

[1]:

True

[2]:

##
## Reconocimiento de entidades en una frase
##
sent = nltk.corpus.treebank.tagged_sents()[22]

##
## El parametro binary=True hace que las entidades
## sean taggeadas unicamente como NE
##
print(nltk.ne_chunk(sent, binary=True))

(S
  The/DT
  (NE U.S./NNP)
  is/VBZ
  one/CD
  of/IN
  the/DT
  few/JJ
  industrialized/VBN
  nations/NNS
  that/WDT
  *T*-7/-NONE-
  does/VBZ
  n't/RB
  have/VB
  a/DT
  higher/JJR
  standard/NN
  of/IN
  regulation/NN
  for/IN
  the/DT
  smooth/JJ
  ,/,
  needle-like/JJ
  fibers/NNS
  such/JJ
  as/IN
  crocidolite/NN
  that/WDT
  *T*-1/-NONE-
  are/VBP
  classified/VBN
  *-5/-NONE-
  as/IN
  amphobiles/NNS
  ,/,
  according/VBG
  to/TO
  (NE Brooke/NNP)
  T./NNP
  Mossman/NNP
  ,/,
  a/DT
  professor/NN
  of/IN
  pathlogy/NN
  at/IN
  the/DT
  (NE University/NNP)
  of/IN
  (NE Vermont/NNP College/NNP)
  of/IN
  (NE Medicine/NNP)
  ./.)

[3]:

##
## El parametro binary=False hace que las entidades
## sean taggeadas adicionando categorias:
##
##  ORGANIZATION
##  LOCATION
##  DATE
##  TIME
##  MONEY
##  PERCENT
##  FACILITY  human-made artifacts in the domains of architecture and civil engineering
##  GPE geo-political entities such as city, state/province, and country.
##
print(nltk.ne_chunk(sent))

(S
  The/DT
  (GPE U.S./NNP)
  is/VBZ
  one/CD
  of/IN
  the/DT
  few/JJ
  industrialized/VBN
  nations/NNS
  that/WDT
  *T*-7/-NONE-
  does/VBZ
  n't/RB
  have/VB
  a/DT
  higher/JJR
  standard/NN
  of/IN
  regulation/NN
  for/IN
  the/DT
  smooth/JJ
  ,/,
  needle-like/JJ
  fibers/NNS
  such/JJ
  as/IN
  crocidolite/NN
  that/WDT
  *T*-1/-NONE-
  are/VBP
  classified/VBN
  *-5/-NONE-
  as/IN
  amphobiles/NNS
  ,/,
  according/VBG
  to/TO
  (PERSON Brooke/NNP T./NNP Mossman/NNP)
  ,/,
  a/DT
  professor/NN
  of/IN
  pathlogy/NN
  at/IN
  the/DT
  (ORGANIZATION University/NNP)
  of/IN
  (PERSON Vermont/NNP College/NNP)
  of/IN
  (GPE Medicine/NNP)
  ./.)

[4]:

##
## Las relaciones son extraidas como una tripleta (X, a, Y)
## donde X y Y son entidades y `a` representa la relación
##
import re
IN = re.compile(r".*\bin\b(?!\b.+ing)")
for doc in nltk.corpus.ieer.parsed_docs("NYT_19980315"):
    for rel in nltk.sem.extract_rels("ORG", "LOC", doc, corpus="ieer", pattern=IN):
        print(nltk.sem.rtuple(rel))

[ORG: 'WHYY'] 'in' [LOC: 'Philadelphia']
[ORG: 'McGlashan & Sarrail'] 'firm in' [LOC: 'San Mateo']
[ORG: 'Freedom Forum'] 'in' [LOC: 'Arlington']
[ORG: 'Brookings Institution'] ', the research group in' [LOC: 'Washington']
[ORG: 'Idealab'] ', a self-described business incubator based in' [LOC: 'Los Angeles']
[ORG: 'Open Text'] ', based in' [LOC: 'Waterloo']
[ORG: 'WGBH'] 'in' [LOC: 'Boston']
[ORG: 'Bastille Opera'] 'in' [LOC: 'Paris']
[ORG: 'Omnicom'] 'in' [LOC: 'New York']
[ORG: 'DDB Needham'] 'in' [LOC: 'New York']
[ORG: 'Kaplan Thaler Group'] 'in' [LOC: 'New York']
[ORG: 'BBDO South'] 'in' [LOC: 'Atlanta']
[ORG: 'Georgia-Pacific'] 'in' [LOC: 'Atlanta']