摘要: … will show, past systems were not evaluated under general conditions and, therefore, the performance reported is not … , we recognize that the set of cases on which a classifier is accurate …
Guido Zarrella, John D. Burger, John Henderson, George Kim, Discriminating Gender on Twitterempirical methods in natural language processing. pp. 1301- 1309 ,(2011)
Jacob Ratkiewicz, Alessandro Flammini, Michael D. Conover, Filippo Menczer, Bruno Gonçalves, Matthew R. Francisco, Political Polarization on Twitterinternational conference on weblogs and social media. ,(2011)