Largest dataset ever of African American English

Sources: ScienceDailyUMass Amherst, Lisa Green and Brendan O’Connor collaborated with doctoral student Su Lin Blodgett on a case study of African American English in on-line Twitter conversations. The authors have created what they believe to be the largest data set of African American English to date, examining 59 million tweets from 2.8 million users. Their goal is to characterize and identify dialects, and to ultimately create language technology that is adapted to African American English.

Facebooktwitterredditmailby feather