def execute_command(self, mecab, message):
fun_talk = []
# TODO : 개선
fun_talk.append('흠..그럼 컬투쇼 사연 하나 소개해 드릴께요ㅋㅋ')
fun_talk.append(PresenTALKUtil.get_random_message('fun_talk'))
fun_talk.append(PresenTALKUtil.get_random_message('laugh'))
fun_talk.append(PresenTALKUtil.get_random_message('apologize'))
return_dict = {'command': 'fun_talk', 'message': fun_talk}
return return_dict
def __init__(self):
# 챗봇 종료
self.isFinish = False
# 형태소 분석기
self.mecab = Mecab()
# config
self.intent_analyzer = intent_analyzer.IntentAnalyzer()
# 사용자 생성
self.user = User()
# 대화 내용 list
self.conversation_list = []
# 토픽 list ( 기본 정의 )
self.topic_list = PresenTALKUtil.get_topic_list()
# 룰 딕셔너리 ( 특정 토픽에 대한 룰 )
self.rule_list = {}
# 룰 리스트 ( 모든 토픽에 대한 룰 )
# Topic 을 찾지 못했을 때는, 모든 rule list 에 대해서 전부 매칭을 진행 -> 이건 추후 변경 여지가 있음 -> Proto Type 에 대해서는 이렇게 진행하고자 함
self.all_rule_list = []
# feature 딕셔너리
# 일단 feature 를 세 개로 지정 -> 추후 늘리거나 추천 알고리즘에 맞게 적용
self.feature_dict = {'object': '', 'age': '', 'purpose': ''}
# feature 딕셔너리 후보 ( 일상대화에서 feature 후보군이 나올수도 있음 )
self.feature_candidate_dict = {'object': '', 'age': '', 'purpose': ''}
# Rule list
self.temp_rule_list = PresenTALKUtil.get_temp_rule_list()
for rule_row in self.temp_rule_list:
topic_id = rule_row['topic_id']
if topic_id in self.rule_list.keys():
temp_rule_dict = {
'matching_rule': rule_row['matching_rule'],
'response_rule': rule_row['response_rule']
}
self.rule_list[topic_id].append(temp_rule_dict)
else:
temp_rule_list = []
temp_rule_dict = {
'matching_rule': rule_row['matching_rule'],
'response_rule': rule_row['response_rule']
}
temp_rule_list.append(temp_rule_dict)
self.rule_list[topic_id] = temp_rule_list
self.all_rule_list.append(temp_rule_dict)
def get_user_info_message(self):
return_message = ''
for key in self.user.user_info.keys():
if self.user.user_info[key] == '':
return_message = PresenTALKUtil.get_random_message(key)
break
return return_message
def print_feature_message(self):
return_feature_dic = {'feature_key': '', 'feature_message': ''}
for key in self.feature_dict.keys():
if self.feature_dict[key] == '':
return_feature_dic['feature_key'] = key
return_feature_dic[
'feature_message'] = PresenTALKUtil.get_random_message(key)
break
return return_feature_dic
def analyze_data(self, file_path):
present_list = PresenTALKUtil.get_all_present_list()
present_target = PresenTALKUtil.get_feature_target()
present_purpose = PresenTALKUtil.get_feature_purpose()
data_file = open(file_path, 'r', encoding='utf-8')
lines = data_file.readlines()
word_count_dict = dict()
for line in lines:
word_list = line.split(' ')
for word in word_list:
word = word.replace('\n', '')
if word in word_count_dict:
word_count_dict[word] += 1
else:
word_count_dict[word] = 1
# word Count
sorted_list = sorted(word_count_dict.items(), key=itemgetter(1), reverse=True)
# for key, value in sorted_list:
# print(key, value)
# present word count
for key, value in sorted_list:
if key in present_list:
print(key, value)
# target word count
for key, value in sorted_list:
if key in present_target:
print(key, value)
# purpose word count
for key, value in sorted_list:
if key in present_purpose:
print(key, value)
def get_is_present(word):
is_present = False;
present_list = PresenTALKUtil.get_present_list()
for present in present_list:
if present in word:
is_present = True
break
else:
if word in wiki_model.wv.vocab:
if wiki_model.similarity('선물', word) > 0.3:
is_present = True
break
return is_present
def find_feature(self, message, feature_key):
# feature 에 대한 정보를 사용자가 이야기했는지를 체크해서 feature 리스트에 담아 두어야 한다
# feature 를 어떻게 찾을 것인가? -> 포함 ? 명사 비교 ?
feature_dic_list = PresenTALKUtil.get_feature_dictionary(feature_key)
# 명사 카운팅을 하자
# 이유 : 명사가 여러 개인 경우, 포함된 feature 가 꼭 그 feature 일 가능성이 점점 낮아짐
# ex) "얼마전에 엄마 선물을 해드려서 아빠도 좀 해드리려고"
is_find = False
for feature in feature_dic_list:
feature_list = feature['concept_list'].split(',')
feature_type = feature['feature_type']
feature_root = feature['concept_name']
for m_feature in feature_list:
if m_feature in message:
is_find = True
self.feature_dict[feature_type] = feature_root
return is_find
def find_feature_candidate(self, message):
# message 의 길이에 따라 로직을 달리 하자
# length 가 길면 여러 의미로 해석될 수 있고, 짧은 경우에는 대부분의 내용이 핵심 내용일 확률이 높다
# 뒤에 불필요한 내용은 제거 후 길이 확인 ( <= 20 )
message = text_util.remove_unsueful_text(message)
if len(message) <= 20:
all_feature_dic_list = PresenTALKUtil.get_feature_dictionary('')
for feature in all_feature_dic_list:
feature_list = feature['concept_list'].split(',')
feature_type = feature['feature_type']
feature_root = feature['concept_name']
for m_feature in feature_list:
if m_feature in message:
self.feature_dict[feature_type] = feature_root
else:
pass
def __init__(self):
self._neo4j_c = neo4j_client.Neo4jClient('your db host', 'your db name', 'your db password')
self._node_list = self._neo4j_c.execute_query('MATCH (n) RETURN n.name, n.alias', ['n.name', 'n.alias'])
self._relation_list = self._neo4j_c.execute_query('MATCH (n)-[r]-() RETURN DISTINCT type(r), r.alias',
['type(r)', 'r.alias'])
self._prop_list = PresenTALKUtil.get_knowledge_prop_list()
def __init__(self):
self.keyword_list = PresenTALKUtil.get_command_config('recommend')[0]['command_word'].split(',')
def execute_fragments(self, mecab, message):
# TODO : feature 를 찾는 로직 & feature 정보가 기준 이상이라면 그 feature 정보를 바탕으로 추천
# 그게 아니라면 일반적인 선물 이야기하기 ( PresenTALKUtil.get_random_message() 활용 )
return PresenTALKUtil.get_random_message('normal_present')
def __init__(self):
self.end_post_list = [
'VX+EC', 'XSV+EC', 'VX', 'EP+EF', 'VX+EF', 'VV+EC', 'VV', 'EC'
]
self.command_config_list = PresenTALKUtil.get_command_config('')
def response_quibble_message(self, message):
# default
response_message = []
response_message.append(PresenTALKUtil.get_random_message('quibble'))
return response_message
def print_gambit_message(self, message, intent_index, message_type,
pre_feature_key):
# TODO : 로직 정리
print(message_type)
response_message_type = ''
feature_key = ''
response_text = []
if message_type == 'BG':
response_text.append(
PresenTALKUtil.get_random_message('bot_starting'))
# BS : Bot Start, BQ : Bot Question
response_message_type = 'BS'
elif message_type == 'BS':
# 사용자의 응답이 긍정일 경우
if BayesClassifier.get_message_class(message) == 'POS':
temp_dic = self.print_feature_message()
feature_key = temp_dic['feature_key']
response_text.append(temp_dic['feature_message'])
response_message_type = 'BF'
# 긍정이 아닐 경우
else:
# 사용자가 원하는 걸 직접 질문
response_text.append(
PresenTALKUtil.get_random_message('reject'))
response_message_type = 'BJ'
elif message_type == 'BF':
is_find = False
if pre_feature_key != '':
# feature 추출
is_find = self.find_feature(message, pre_feature_key)
temp_dic = self.print_feature_message()
feature_key = temp_dic['feature_key']
# feature 추출이 모두 끝난 경우
if feature_key == '':
# rec = recommand_present.Recommandation()
# response_text.append(rec.get_recommand_message(self.feature_dict))
# who, age, price, why
recommend_list = recommend_data2.recommend_in_DB(
self.feature_dict['object'], self.feature_dict['age'],
'10', self.feature_dict['purpose'])
present_list = ''
for present in recommend_list:
if present_list == '':
present_list = present
else:
present_list = present_list + ', ' + present
recommend_text = '선물 추천 결과입니다. ' + str(
present_list) + ' 어떠신가요?'
response_text.append(recommend_text)
response_message_type = 'BR'
else:
response_text.append(temp_dic['feature_message'])
response_message_type = 'BF'
# TODO : 추후 추가 ( 로직의 복잡도 증가 )
# elif message_type == 'BI':
# if BayesClassifier.get_message_class(message) == 'POS':
# pass
# else:
# pass
elif message_type == 'BN':
response_text = self.find_matching_rule(message)
if len(response_text) == 0:
response_message_type = 'BN'
response_text = self.response_quibble_message(message)
# TODO : 추후 추가 ( 로직의 복잡도 증가 )
# 매칭하는 룰을 찾지 못했을 경우 특정 토픽으로 이야기하기
# if len(response_text) == 0:
# user_info_text = self.get_user_info_message()
#
# if user_info_text == '':
# response_message_type = 'BN'
# response_text = self.response_quibble_message(message)
# else:
# response_text.append('제가 공부한 분야로 이야기해 보시겠어요?')
# response_text.append(user_info_text)
# response_message_type = 'BI'
elif message_type == 'BR':
# TODO : 로직 정리 필요
if BayesClassifier.get_message_class(message) == 'POS':
response_message_type = 'BE'
response_text.append(
PresenTALKUtil.get_random_message('bot_ending'))
else:
# 사용자가 원하는 걸 직접 질문
response_text.append(
PresenTALKUtil.get_random_message('reject'))
response_message_type = 'BJ'
# TODO : 대답 하고 나서의 로직 추가
elif message_type == 'BA':
return_message = self.find_matching_rule(message)
if len(return_message) != 0:
response_text = return_message
else:
response_text = self.response_quibble_message(message)
response_message_type = 'BN'
elif message_type == 'BJ':
if BayesClassifier.get_message_class(message) == 'POS':
response_message_type = 'BE'
response_text.append(
PresenTALKUtil.get_random_message('bot_ending'))
else:
response_message_type = 'BN'
response_text.append(
PresenTALKUtil.get_random_message('bot_restart'))
elif message_type == 'BE':
if BayesClassifier.get_message_class(message) == 'POS':
self.isFinish = True
response_text.append(
PresenTALKUtil.get_random_message('bot_bye'))
# TODO : 시나리오 구성
else:
return_message = self.find_matching_rule(message)
if len(return_message) != 0:
response_text = return_message
else:
response_text = self.response_quibble_message(message)
response_message_type = 'BN'
self.save_conversation_list(response_text, response_message_type,
feature_key)
return response_text
def print_greeting_message(self):
greeting_message = []
greeting_message.append(
PresenTALKUtil.get_random_message('bot_greeting'))
self.save_conversation_list(greeting_message, 'BG', '')
return greeting_message