We should add an Ngrams context to ContextExtractor that allows you to extract
all n-grams of features. For example, a document classifier using token bigrams
as features could do something like:
new ContextExtractor(Token.class, coveredTextExtractor, new Ngrams(2, new Covered()));
Original issue reported on code.google.com by steven.b...@gmail.com on 7 Apr 2012 at 8:31
Original issue reported on code.google.com by
steven.b...@gmail.com
on 7 Apr 2012 at 8:31