def get_f_contain_name_without_last(self):
f_contain_name = 0
last_name = self.person_name.split(' ')[-1]
name_part_list = self.person_name.split(' ')[:-1]
# 将名字部分列表按长度排序，长的部分先参与计算
name_part_list = sorted(name_part_list, key=lambda name_part: len(name_part), reverse=True)
temp_addr = self.email_addr[:self.email_addr.find('@')].replace('.', '').replace('_', '').replace('-', '')
temp_addr = del_a_from_b(last_name, temp_addr)
for name_part in name_part_list:
# 如果名字的某部分在前缀中，f_contain_name加上该部分的长度，并从前缀和name_part_list中去掉该部分
if name_part in temp_addr:
temp_addr = del_a_from_b(name_part, temp_addr)
f_contain_name += len(name_part)
# 在for循环中使用remove是危险的，故标为空，相当于删除这个名字部分
name_part_list[name_part_list.index(name_part)] = ''
# 对于没有被全包含的名字部分，看其首字母是否在前缀中
for name_part in name_part_list:
if name_part and name_part[0] in temp_addr:
f_contain_name += 1
# 归一化
f_contain_name = float(f_contain_name) / (float(len(del_a_from_b(last_name, self.email_addr[:self.email_addr.find('@')]))) + 1)
return f_contain_name
def get_f_contain_name_without_last(self):
f_contain_name = 0
last_name = self.person_name.split(' ')[-1]
name_part_list = self.person_name.split(' ')[:-1]
# 将名字部分列表按长度排序，长的部分先参与计算
name_part_list = sorted(name_part_list,
key=lambda name_part: len(name_part),
reverse=True)
temp_addr = self.email_addr[:self.email_addr.find('@')].replace(
'.', '').replace('_', '').replace('-', '')
temp_addr = del_a_from_b(last_name, temp_addr)
for name_part in name_part_list:
# 如果名字的某部分在前缀中，f_contain_name加上该部分的长度，并从前缀和name_part_list中去掉该部分
if name_part in temp_addr:
temp_addr = del_a_from_b(name_part, temp_addr)
f_contain_name += len(name_part)
# 在for循环中使用remove是危险的，故标为空，相当于删除这个名字部分
name_part_list[name_part_list.index(name_part)] = ''
# 对于没有被全包含的名字部分，看其首字母是否在前缀中
for name_part in name_part_list:
if name_part and name_part[0] in temp_addr:
f_contain_name += 1
# 归一化
f_contain_name = float(f_contain_name) / (float(
len(
del_a_from_b(last_name,
self.email_addr[:self.email_addr.find('@')]))) +
1)
return f_contain_name
def get_f_first_char_all_in_addr(self):
name_part_list = self.person_name.split(' ')
first_char_in_num = 0
email_prefix = self.email_addr[:self.email_addr.find('@')]
prefix_len = len(email_prefix)
for name_part in name_part_list:
if name_part[0] in email_prefix:
first_char_in_num += 1
del_a_from_b(name_part[0], email_prefix)
return float(first_char_in_num) / float(prefix_len)
def get_f_first_char_all_in_addr(self):
name_part_list = self.person_name.split(' ')
first_char_in_num = 0
email_prefix = self.email_addr[:self.email_addr.find('@')]
prefix_len = len(email_prefix)
for name_part in name_part_list:
if name_part[0] in email_prefix:
first_char_in_num += 1
del_a_from_b(name_part[0], email_prefix)
return float(first_char_in_num) / float(prefix_len)
def get_binary_relationship_list(self):
binary_relationship_list = []
for node_looper in range(len(self.node_mln_list)):
node = self.node_mln_list[node_looper]
for another_node_looper in range(node_looper+1, len(self.node_mln_list)):
another_node = self.node_mln_list[another_node_looper]
if node.node_name == another_node.node_name:
continue
# if node.prefix == another_node.prefix and node.domain != another_node.domain and not node.prefix_is_invalid_keyword()[0]:
# binary_relationship_list.append(node.grounding_string_binary('same_prefix', another_node.node_name))
if node.domain == another_node.domain and node.prefix == another_node.prefix:
binary_relationship_list.append(node.grounding_string_binary('same_addr', another_node.node_name))
if (node.domain == another_node.domain or node.domain in another_node.domain or another_node.domain in node.domain) and node.prefix != another_node.prefix and another_node.prefix_is_invalid_keyword()[0]:
binary_relationship_list.append(node.grounding_string_binary('same_domain_with_invalid', another_node.node_name))
if node.prefix != another_node.prefix and (another_node.prefix in node.prefix or node.prefix in another_node.prefix):
big_node = node
small_node = another_node
if node.prefix in another_node.prefix:
big_node = another_node
small_node = node
remain_prefix = del_a_from_b(small_node.prefix, big_node.prefix)
if remain_prefix[0] in big_node.first_char_list:
binary_relationship_list.append(node.grounding_string_binary('a_contain_prefix_b', another_node.node_name))
if node.prefix == another_node.prefix and node.domain != another_node.domain and not node.prefix_is_invalid_keyword()[0]:
binary_relationship_list.append(node.grounding_string_binary('same_prefix', another_node.node_name))
if node.addr_repeat_time > 2:
binary_relationship_list.append(node.grounding_string_binary('addr_repeat_over_3', node.node_name))
# elif node.addr_repeat_time > 0:
# binary_relationship_list.append(node.grounding_string_binary('addr_repeat_under_2', node.node_name))
return binary_relationship_list
def get_binary_relationship_list(self):
binary_relationship_list = []
for node_looper in range(len(self.node_mln_list)):
node = self.node_mln_list[node_looper]
for another_node_looper in range(node_looper + 1,
len(self.node_mln_list)):
another_node = self.node_mln_list[another_node_looper]
if node.node_name == another_node.node_name:
continue
# if node.prefix == another_node.prefix and node.domain != another_node.domain and not node.prefix_is_invalid_keyword()[0]:
# binary_relationship_list.append(node.grounding_string_binary('same_prefix', another_node.node_name))
if node.domain == another_node.domain and node.prefix == another_node.prefix:
binary_relationship_list.append(
node.grounding_string_binary('same_addr',
another_node.node_name))
if (
node.domain == another_node.domain
or node.domain in another_node.domain
or another_node.domain in node.domain
) and node.prefix != another_node.prefix and another_node.prefix_is_invalid_keyword(
)[0]:
binary_relationship_list.append(
node.grounding_string_binary(
'same_domain_with_invalid',
another_node.node_name))
if node.prefix != another_node.prefix and (
another_node.prefix in node.prefix
or node.prefix in another_node.prefix):
big_node = node
small_node = another_node
if node.prefix in another_node.prefix:
big_node = another_node
small_node = node
remain_prefix = del_a_from_b(small_node.prefix,
big_node.prefix)
if remain_prefix[0] in big_node.first_char_list:
binary_relationship_list.append(
node.grounding_string_binary(
'a_contain_prefix_b', another_node.node_name))
if node.prefix == another_node.prefix and node.domain != another_node.domain and not node.prefix_is_invalid_keyword(
)[0]:
binary_relationship_list.append(
node.grounding_string_binary('same_prefix',
another_node.node_name))
if node.addr_repeat_time > 2:
binary_relationship_list.append(
node.grounding_string_binary('addr_repeat_over_3',
node.node_name))
# elif node.addr_repeat_time > 0:
# binary_relationship_list.append(node.grounding_string_binary('addr_repeat_under_2', node.node_name))
return binary_relationship_list
