Files
twitter_scraper/code/clean_data.py
T
2023-01-09 19:03:56 +01:00

32 lines
918 B
Python

# public packages
from cleantext import clean
def clean_content(text):
return clean(text, to_ascii=False) # avoid ascii, as it does not have æ,ø,å
def clean_content_df(df):
content_clean_list = df['content'].apply(clean_content).to_list()
id_list = df['id'].to_list()
return content_clean_list, id_list
def clean_content_en(text_list):
text_list_clean = list()
for text in text_list:
text_clean = clean(
text=text,
fix_unicode=True,
to_ascii=True,
lower=True,
no_line_breaks=True,
no_urls=True,
no_emails=True,
no_phone_numbers=True,
no_numbers=True,
no_digits=True,
no_currency_symbols=True,
no_punct=True,
no_emoji=True,
lang='en'
)
text_list_clean.append(text_clean)
return text_list_clean