from entity_verb.nlp import NLP
import json
f =open('entity_verb_result\\' + "all_entity.json"
, 'r', encoding='utf-8')
file = f.read()
all_entity = json.loads(file)['all_entity']
f.close()
nlp =NLP()
postag_dict = dict()
for word in all_entity:
postage = nlp.get_postag(word)
if postage not in postag_dict.keys():
postag_dict[postage] = [word]
else:
postag_dict[postage].append(word)
print(postag_dict)
for i in postag_dict.keys():
postag_dict_new = dict()
postag_dict_new[i+'_'+str(len(postag_dict[i]))] = list(postag_dict[i])
with open("entity_verb_result\\entity_classification_LTP.json", 'a') as f_out:
f_out.write(json.dumps(postag_dict_new,ensure_ascii=False))
f_out.write("\n")
# print(all_entity)
if __name__ == "__main__":
# 读取文件
entity_verb_new = entity_verb_new()
"""
加载LTP的分词器和词性标注器
"""
default_model_dir = 'D:\python-file\knowledge_extraction-master-tyz\\ltp_data_v3.4.0\\' # LTP模型文件目录
segmentor = Segmentor()
user_dict = "source\\user.txt"
segmentor_flag = segmentor.load_with_lexicon(
os.path.join(default_model_dir, 'cws.model'), user_dict)
postagger = Postagger()
postag_flag = postagger.load(os.path.join(default_model_dir, 'pos.model'))
nlp = NLP()
thu1 = thulac.thulac() # 默认模式
path = r"D:\python-file\北京市旅游知识图谱\\verb-entity\\bj_travel"
file_list = os.listdir(path)
f = open('entity_verb_result\\' + "all_entity.json", 'r', encoding='utf-8')
file = f.read()
all_entity = json.loads(file)['all_entity']
print(all_entity)
f.close()
for file_name in file_list:
print(file_name)
f = open('D:\python-file\北京市旅游知识图谱\\verb-entity\\bj_travel\\' +
file_name,
'r',
encoding='utf-8')