Closed tupini07 closed 5 years ago
Plain URL tokens might not be good as a feature since they have a lot of noise. Consider using the list of url stopwords found in soweego.commons.resources.urls_stop_words.txt so that only relevant tokens are considered for comparison.
url stopwords
soweego.commons.resources.urls_stop_words.txt
Plain URL tokens might not be good as a feature since they have a lot of noise. Consider using the list of
url stopwords
found insoweego.commons.resources.urls_stop_words.txt
so that only relevant tokens are considered for comparison.