Are you considering punctuations as “Other”. I thought we were going to remove punctuation from the text, as stopwords.
To have them in or out changes the transition matrix. Are they “stopwords” or not?
Are you considering punctuations as “Other”. I thought we were going to remove punctuation from the text, as stopwords.
To have them in or out changes the transition matrix. Are they “stopwords” or not?
In this picture - yes, the “.” and “:” are considered as “O”.
In the assignment :
'' # $ ( ) , . : ``
are different categories (9 out of 46 total). We do not consider them as stopwords, because they might help determine the POS.
Using stopwords or not is a design choice - depending on your application they might be needed or not.
Cheers