2007 NAACL NAACL 2007

Using “Annotator Rationales” to Improve Machine Learning for Text Categorization