def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for chunk in sentence:
if '動詞' in [m.pos for m in chunk.morphs]:
predicate = [
m.base for m in chunk.morphs if m.pos == '動詞'
][0]
arguments = []
for src in chunk.srcs:
if '助詞' in [m.pos for m in sentence[src].morphs]:
argument = sentence[src].get_text()
case = [
m.base for m in sentence[src].morphs
if m.pos == '助詞'
][-1]
arguments.append((case, argument))
if arguments:
arguments.sort()
print '{0}\t{1}\t{2}'.format(
predicate, ' '.join([a[0] for a in arguments]),
' '.join([a[1] for a in arguments]))
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for i in range(len(sentence)-1):
for j in range(i+1, len(sentence)):
if '名詞' in [m.pos for m in sentence[i].morphs] and '名詞' in [m.pos for m in sentence[j].morphs]:
# Get indices of morphs to be replaced by 'X' or 'Y'
index_x = []
index_x_flag = False
for index, m in enumerate(sentence[i].morphs):
if m.pos == '名詞':
index_x_flag = True
index_x.append(index)
else:
if index_x_flag:
break
index_y = []
index_y_flag = False
for index, m in enumerate(sentence[j].morphs):
if m.pos == '名詞':
index_y_flag = True
index_y.append(index)
else:
if index_y_flag:
break
# Get paths of syntax tree
path_i = []
path_j = []
node = i
while True:
path_i.append(node)
node = sentence[node].dst
if node is None:
break
node = j
while True:
path_j.append(node)
node = sentence[node].dst
if node is None:
break
# Display syntax tree
if set(path_i) >= set(path_j):
path_texts = self.get_replaced_texts(sentence, sorted(list(set(path_i) - set(path_j)) + [j]), i, j, index_x, index_y)
print ' -> '.join(path_texts)
else:
path_i_only_texts = self.get_replaced_texts(sentence, sorted(list(set(path_i) - set(path_j))), i, j, index_x, index_y)
path_j_only_texts = self.get_replaced_texts(sentence, sorted(list(set(path_j) - set(path_i))), i, j, index_x, index_y)
path_common_texts = self.get_replaced_texts(sentence, sorted(list(set(path_i) & set(path_j))), i, j, index_x, index_y)
print '{0} | {1} | {2}'.format(' -> '.join(path_i_only_texts), ' -> '.join(path_j_only_texts), ' -> '.join(path_common_texts))
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
print text[7]
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
print text[7]
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for chunk in sentence:
if chunk.dst is not None:
print '{0}\t{1}'.format(chunk.get_text(), sentence[chunk.dst].get_text())
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
sentence = text[7] # As the dependency graph becomes too complex, we visualize only 8th sentence in this time
edges = []
for chunk in sentence:
if chunk.dst is not None:
edges.append((chunk.get_text(), sentence[chunk.dst].get_text()))
g = pydot.graph_from_edges(edges)
g.write_jpeg('44.jpg', prog='dot')
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
sentence = text[
7] # As the dependency graph becomes too complex, we visualize only 8th sentence in this time
edges = []
for chunk in sentence:
if chunk.dst is not None:
edges.append(
(chunk.get_text(), sentence[chunk.dst].get_text()))
g = pydot.graph_from_edges(edges)
g.write_jpeg('44.jpg', prog='dot')
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for index, chunk in enumerate(sentence):
if '名詞' in [m.pos for m in chunk.morphs]:
path = []
node = index
while True:
path.append(sentence[node].get_text())
node = sentence[node].dst
if node is None:
break
print ' -> '.join(path)
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for index, chunk in enumerate(sentence):
if '名詞' in [m.pos for m in chunk.morphs]:
path = []
node = index
while True:
path.append(sentence[node].get_text())
node = sentence[node].dst
if node is None:
break
print ' -> '.join(path)
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for j in range(len(sentence) - 1):
for k in range(len(sentence[j].morphs) - 1):
if sentence[j].morphs[k].pos == '名詞' and sentence[
j].morphs[k].pos1 == 'サ変接続' and sentence[j].morphs[
k + 1].pos == '助詞' and sentence[j].morphs[
k + 1].surface == 'を' and '動詞' in [
m.pos for m in sentence[
sentence[j].dst].morphs
]:
predicate = sentence[j].get_text() + [
m.base for m in sentence[sentence[j].dst].morphs
if m.pos == '動詞'
][0]
arguments = []
for src in list(
set(sentence[j].srcs +
sentence[sentence[j].dst].srcs) -
set([j])):
if '助詞' in [m.pos for m in sentence[src].morphs]:
argument = sentence[src].get_text()
case = [
m.base for m in sentence[src].morphs
if m.pos == '助詞'
][-1]
arguments.append((case, argument))
if arguments:
arguments.sort()
print '{0}\t{1}\t{2}'.format(
predicate, ' '.join([a[0] for a in arguments]),
' '.join([a[1] for a in arguments]))
break
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for chunk in sentence:
if '動詞' in [m.pos for m in chunk.morphs]:
predicate = [m.base for m in chunk.morphs if m.pos == '動詞'][0]
arguments = []
for src in chunk.srcs:
if '助詞' in [m.pos for m in sentence[src].morphs]:
argument = sentence[src].get_text()
case = [m.base for m in sentence[src].morphs if m.pos == '助詞'][-1]
arguments.append((case, argument))
if arguments:
arguments.sort()
print '{0}\t{1}\t{2}'.format(predicate, ' '.join([a[0] for a in arguments]), ' '.join([a[1] for a in arguments]))
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for j in range(len(sentence)-1):
for k in range(len(sentence[j].morphs)-1):
if sentence[j].morphs[k].pos == '名詞' and sentence[j].morphs[k].pos1 == 'サ変接続' and sentence[j].morphs[k+1].pos == '助詞' and sentence[j].morphs[k+1].surface == 'を' and '動詞' in [m.pos for m in sentence[sentence[j].dst].morphs]:
predicate = sentence[j].get_text() + [m.base for m in sentence[sentence[j].dst].morphs if m.pos == '動詞'][0]
arguments = []
for src in list(set(sentence[j].srcs + sentence[sentence[j].dst].srcs) - set([j])):
if '助詞' in [m.pos for m in sentence[src].morphs]:
argument = sentence[src].get_text()
case = [m.base for m in sentence[src].morphs if m.pos == '助詞'][-1]
arguments.append((case, argument))
if arguments:
arguments.sort()
print '{0}\t{1}\t{2}'.format(predicate, ' '.join([a[0] for a in arguments]), ' '.join([a[1] for a in arguments]))
break
return None
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for chunk in sentence:
if '動詞' in [m.pos for m in chunk.morphs]:
predicate = [
m.base for m in chunk.morphs if m.pos == '動詞'
][0]
cases = []
for src in chunk.srcs:
if '助詞' in [m.pos for m in sentence[src].morphs]:
case = [
m.base for m in sentence[src].morphs
if m.pos == '助詞'
][-1]
cases.append(case)
if cases:
cases.sort()
print '{0}\t{1}'.format(predicate, ' '.join(cases))
def solve(self):
'''
insert your code
'''
pcttc = ParseCabochaTextToChunks()
text = pcttc.parse('neko.txt.cabocha')
for sentence in text:
for i in range(len(sentence) - 1):
for j in range(i + 1, len(sentence)):
if '名詞' in [
m.pos for m in sentence[i].morphs
] and '名詞' in [m.pos for m in sentence[j].morphs]:
# Get indices of morphs to be replaced by 'X' or 'Y'
index_x = []
index_x_flag = False
for index, m in enumerate(sentence[i].morphs):
if m.pos == '名詞':
index_x_flag = True
index_x.append(index)
else:
if index_x_flag:
break
index_y = []
index_y_flag = False
for index, m in enumerate(sentence[j].morphs):
if m.pos == '名詞':
index_y_flag = True
index_y.append(index)
else:
if index_y_flag:
break
# Get paths of syntax tree
path_i = []
path_j = []
node = i
while True:
path_i.append(node)
node = sentence[node].dst
if node is None:
break
node = j
while True:
path_j.append(node)
node = sentence[node].dst
if node is None:
break
# Display syntax tree
if set(path_i) >= set(path_j):
path_texts = self.get_replaced_texts(
sentence,
sorted(list(set(path_i) - set(path_j)) + [j]),
i, j, index_x, index_y)
print ' -> '.join(path_texts)
else:
path_i_only_texts = self.get_replaced_texts(
sentence,
sorted(list(set(path_i) - set(path_j))), i, j,
index_x, index_y)
path_j_only_texts = self.get_replaced_texts(
sentence,
sorted(list(set(path_j) - set(path_i))), i, j,
index_x, index_y)
path_common_texts = self.get_replaced_texts(
sentence,
sorted(list(set(path_i) & set(path_j))), i, j,
index_x, index_y)
print '{0} | {1} | {2}'.format(
' -> '.join(path_i_only_texts),
' -> '.join(path_j_only_texts),
' -> '.join(path_common_texts))
return None