Python文本中选取地点位置

1.固定格式求地点

location_str = ['长沙开区橄榄城小区']#data_zhuti_qumin_tq[1]#投诉长沙开区橄榄城小区车位违建#["徐汇区虹漕路461号58号楼5楼", "泉州市洛江区万安塘西工业区", "朝阳区北苑华贸城"]
import addressparser
df = addressparser.transform(location_str)
print(df)

结果
2.一段文本求地点

import paddlehub as hub
lac = hub.Module(name='lac')
test_text=['投诉长沙开区橄榄城小区车位违建']#,'衡阳县关市网吧常年收留未成年人上网']
inputs = {'text':test_text}
results = lac.lexical_analysis(data = inputs)
for result in results:
    print(result['word'])
    print(result['tag'])

loc = ''
for i in range(len(result['tag'])):
    if result['tag'][i] == 'LOC':
        loc += result['word'][i]
print(loc)

3.正则表达式相似文本求地点

import re
import paddlehub as hub
import addressparser
lac = hub.Module(name='lac')

def location(id): #输入问题ID可以求特定人群地点
    loc_sum = '' #主题里抽取出所有提取的地点，后边识别会更完善
    true = biaoge2['ID'] == id
    for i in range(len(biaoge2[true].head())):
        test_text=[list(biaoge2[true]['主题'])[i]]#,'网吧常年收留未成年人上网']
        inputs = {'text':test_text}
        results = lac.lexical_analysis(data = inputs)
        for result in results:
            loc = ''
            for i in range(len(result['tag'])):
                if result['tag'][i] == 'LOC':# or 'ORG':
                    loc += result['word'][i]
            #print(loc)
                    loc_sum += loc
        loc_str = [loc_sum]
        loc_str_split = addressparser.transform(loc_str)
        
        #对具体地址的处理
        loc_str_split_dizhi = re.sub('([^市]+市|[^县]+县)','',str(list(loc_str_split['地址'])))  #删除市和县的内容
        loc_str_split_dizhi_list = list(set(re.findall('([^镇]+镇|[^村]+村|[^区]+区|[^街]+街)',loc_str_split_dizhi)))  #提取具体坐标
        loc_str_split_dizhi_easy = ''
        for j in range(len(loc_str_split_dizhi_list)): #连接具体地址
            loc_str_split_dizhi_easy += loc_str_split_dizhi_list[j]
        loc_str_split['地址'] = loc_str_split_dizhi_easy
        location =list(loc_str_split['市']+loc_str_split['区']+loc_str_split['地址'])[0]
        
    return location
#             print(result['word'])
#             print(result['tag'])
location(1)