从 pandas json 文件中提取数据

Question

我有以下 json 数据文件，我已将其转换为 pandas 数据帧。栏目如下

Index(['id', 'title', 'abstract', 'content', 'metadata'], dtype='object')

我对'metadata'列特别感兴趣，该列的一个元素看起来像

df_json.loc[78, 'metadata']
"{'classification': {'name': 'Manufacturing, Transport & Logistics'}, 'subClassification': {'name': 'Warehousing, Storage & Distribution'}, 'area': {'name': 'Southern Suburbs & Logan'}, 'location': {'name': 'Brisbane'}, 'suburb': {'name': 'Milton'}, 'workType': {'name': 'Casual/Vacation'}}"

所以我想制作从 'metadata' 列中提取信息的列，例如位置。我不确定如何提取它并将其放在同一个 json 文件旁边，并添加了位置等列

    id  title   abstract    content metadata    clean_content
0   38915469    Recruitment Consultant  We are looking for someone to focus purely on ...   <HTML><p>Are you looking to join a thriving bu...   {'standout': {'bullet1': 'Join a Sector that i...   Are you looking to join a thriving business th...
1   38934839    Computers Salesperson - Coburg  Passionate about exceptional customer service?...   <HTML><p>&middot;&nbsp;&nbsp;Casual hours as r...   {'additionalSalaryText': 'Attractive Commissio...   middotnbspnbspCasual hours as required transit...
2   38946054    Senior Developer | SA   Readifarians are known for discovering the lat...   <HTML><p>Readify helps organizations

Answer 1

 you can use pandas.json_normalize

正在应用您的字符串

 pd.json_normalize(eval(json_string)) 

 #o/p

如果这对你有用，那么你可以试试

 df["metadata"].apply(lambda x: pd.json_normalize(eval(x)))

从 pandas json 文件中提取数据

Extraction from data from pandas json file

python

json

nlp

dataframe