Preprocessing Email Addresses

Read this is you want to know what the preprocessing of email adresses does and what it is useful for

R
Written by Raul Garreta
Updated over a week ago

When the Preprocess email addresses parameter is selected, all email addresses will be replaced by a special word  __email__ . This will allow a model to learn about email mentions in general rather than from specific email mentions. 

This is particularly useful for excluding email addresses from features (i.e.: using the Preprocess email addresses parameter in combination with the model's stopwords) in cases for which the information encoded in emails is of little value for classification purposes. 

Did this answer your question?