[PYTHON] 100 Sprachverarbeitungsklopfen (2020): 31

"""
31.Verb
Extrahieren Sie alle Oberflächenformen des Verbs.

[[{'surface': '', 'base': '*', 'pos': 'BOS/EOS', 'pos1': '*'},
  {'surface': 'einer', 'base': 'einer', 'pos': 'Substantiv', 'pos1': 'Nummer'},
  {'surface': '', 'base': '*', 'pos': 'BOS/EOS', 'pos1': '*'}],
 [{'surface': '', 'base': '*', 'pos': 'BOS/EOS', 'pos1': '*'},
  {'surface': 'ich', 'base': 'ich', 'pos': 'Substantiv', 'pos1': '代Substantiv'},
  {'surface': 'Ist', 'base': 'Ist', 'pos': 'Partikel', 'pos1': '係Partikel'},
  {'surface': 'Katze', 'base': 'Katze', 'pos': 'Substantiv', 'pos1': 'Allgemeines'},
  {'surface': 'damit', 'base': 'Ist', 'pos': 'Hilfsverb', 'pos1': '*'},
  {'surface': 'Gibt es', 'base': 'Gibt es', 'pos': 'Hilfsverb', 'pos1': '*'},
  {'surface': '。', 'base': '。', 'pos': 'Symbol', 'pos1': 'Phrase'},
  {'surface': '', 'base': '*', 'pos': 'BOS/EOS', 'pos1': '*'}],
"""

import itertools
from typing import List

import utils


def get_verb(sentence: List[dict]) -> List:
    result = []
    for word in sentence:
        if word["pos"] == "Verb":
            result.append(word["surface"])
    return result


data = utils.read_json("30_neko_mecab.json")
verbs = [get_verb(sentence) for sentence in data if get_verb(sentence)]
# [['Geboren', 'Tsuka'], ['Shi', 'Weinen', 'Shi', 'Ist'], ['Start', 'Sie sehen'], ['Hör mal zu'], ['Fang', 'Gekocht', 'Essen']]
flat = list(itertools.chain(*verbs))
#  ['Geboren', 'Tsuka', 'Shi', 'Weinen', 'Shi']

Recommended Posts