This algorithm classifies emails and telephones by levels according to different aspects.
The use is simple. Just call the function and declare the parameters.
clas_emails(df,['EMAIL_1','EMAIL_2'])
Where df is the dataframe, EMAIL_1 and EMAIL_2 are the columns to be classified.
A new column is created with the levels that correspond to each email.
In the case of telephones, it works in a similar way.
PARTE 3 and PARTE 7 contain different commented lines. This is because there are different regular expression patterns capable of detecting an email as "invalid". For example, with a regular expression brian-95smith@yahoo.com is a invalid email but in others it's valid. The same thing happens with david88/_s1@company.org which in some cases is invalid and in others valid. It could even depend on the domain as well. That's why I put two more expressions, to show how it identifies one email as invalido and other doesn't.