The U.S. Department of Justice announced the $800,000 project being developed at Cardiff University’s Social Data Science Lab, reports the Financial Times. The project aims to develop an algorithm that scans Twitter to automatically identify hate crime-related tweets in specific regions of the U.S. and compare them to offline hate crimes being committed in the same geographic regions. The project will continue over the next three years and will first be tested by the city of Los Angeles.
“Say a black person was killed by police in the U.S.—and that happens a lot more than it should—we will see biased tweets coming out, using phrases like ‘They had it coming’ or ‘Get them out’,” Peter Burnap, co-lead of the project told the FT. “It doesn’t always have to use derogatory words associated with racism: It could be much more nuanced, which is the major challenge in the project. We are using natural language processing to identify cyber-hate in all its forms.”