verbframedict[verbbase].append(
(adp, chunks[i].show_only_words()))
noun = set()
verb = set()
return verbframedict
if __name__ == "__main__":
with open("ai.ja1.txt.parsed", "r") as ai:
ai = ai.readlines()
ai_morphs = []
for i in range(len(ai)):
ai_morphs.append(Morph(ai[i]))
sents = morph2sents(ai_morphs)
frames = defaultdict(list)
for sent in sents:
dep = morph2chunk(sent)
verbframe2(dep, frames)
#コーパス中で頻出する述語と格パターンの組み合わせ
for key, value in sorted(frames.items(),
key=lambda x: len(x[1]),
reverse=True)[:10]:
output = defaultdict(list)
for elem in value:
output[elem[0]].append(elem[1])
print(key, end="\t")
print(" ".join(output.keys()), end="\t")
print(" ".join([" ".join(elem) for elem in output.values()]))
from collections import defaultdict
from knock40 import Morph, morph2sents
from knock41 import Chunk, morph2chunk, sources
if __name__ == "__main__":
with open("ai.ja1.txt.parsed", "r") as ai:
ai = ai.readlines()
ai_morphs = []
for i in range(len(ai)):
ai_morphs.append(Morph(ai[i]))
sents = morph2sents(ai_morphs)
for sent in sents:
chunks = morph2chunk(sent)
sources(chunks)
for i in range(len(chunks)):
#print(chunk.meta)
#print(chunk.show_bunsetsu_tag())
if "サ変接続" in chunks[i].show_morph_pos1():
if "を" in chunks[i].show_only_listwords():
print(chunks[i].show_only_words(), end="")
goto = chunks[i].dst
verb = chunks[goto].show_base_for_X("動詞")
adpos = set()
dep = []
if verb != None:
if chunks[goto].srcs != []:
for head_id in chunks[goto].srcs:
adp = chunks[head_id].show_base_for_X("助詞")
if adp != None and head_id != i:
def noun2verb(chunks):
noun = set()
verb = set()
for i in range(len(chunks)):
for morph in chunks[i].morphs:
noun.add(morph.pos)
if "名詞" in noun:
goto = chunks[i].dst
for morph2 in chunks[goto].morphs:
verb.add(morph2.pos)
if "動詞" in verb:
print(chunks[i].show_only_words(), end="\t")
print(chunks[goto].show_only_words())
noun = set()
verb = set()
if __name__ == "__main__":
with open("ai.ja1.txt.parsed", "r") as ai:
ai = ai.readlines()
ai_morphs = []
for i in range(len(ai)):
ai_morphs.append(Morph(ai[i]))
dep = morph2chunk(morph2sents(ai_morphs)[1])
noun2verb(dep)