Cardiff project uses Twitter analysis to predict Los Angeles hate crime

During a three-year experiment, researchers from the UK’s Cardiff University will be closely monitoring a massive amount of Tweets related to the Los Angeles area in an attempt to identify patterns and indications that prejudice-motivated violence is about to occur. The scientists will then compare the data against records of actual reported violence.

Eventually, they hope to be able to predict in real time when and where hate crime is likely to take place, and enable policemen to take timely action.

“Developing a better understanding of hateful sentiments online and their relationship with crime on the streets could push law enforcement to better identify, report and address hate crimes that are occurring offline,” said Professor Matt Williams, Director of the Cardiff University’s Social Data Science Lab.

“The insights provided by our work will help US localities to design policies to address specific hate crime issues unique to their jurisdiction and allow service providers to tailor their services to the needs of victims, especially if those victims are members of an emerging category of hate crime targets.”

The project has received $800,000 (£616,200) of funding from the US Department of Justice. It is the first attempt to use social media in the US to create predictive policing models of hate crime. Previously, the LA Police Department successfully experimented with mathematical models to predict other areas of crime. The technology reportedly enabled the police force to reduce crime rates.

“Predictive policing is a proactive law enforcement model that has become more common partially due to the advent of advanced analytics such as data mining and machine-learning methods,” explained Pete Burnap, from Cardiff University’s School of Computer Science & Informatics.

“New analytic approaches and the ability to process very large data sets have increased the accuracy of predictive models over traditional crime analysis methods and this project will evaluate if police departments can leverage these new data and techniques to reduce hate crimes.”

According to the US Bureau of Justice Statistics (BJS), in 2012 an estimated 293,800 incidents of nonfatal violent and property hate crime took place in the US. Prejudice-motivated hate crime involves cases when a victim is being targeted because of his or her affiliation to a social group, such as their sex, ethnicity, disability or religion.

UK official data shows that there were 52,528 hate crimes recorded by the police in England and Wales in 2014/15, an increase of 18 per cent compared with 2013/14.

Previous research from the Social Data Science Lab has shown that Twitter data can be used to identify hot spots, such as certain states or cities, where hate speech has occurred but where hate crime has not been reported. One example is an area where recent immigrants may be unlikely to report crime due to fear of deportation.