When a Twitter algorithm detects riots, violence faster than police; more about the Cardiff University study

RESEARCHERS in UK’s Wales have come out with a Twitter-monitoring algorithm that can identify riots even before citizens actually report about those unrest to police, according to media reports published in UK’s newspapers and elsewhere. Reports say, the model has recently been tested on the basis of data gathered from Twitter during 2011 riots occurred in the city of London. That the algorithm could identify…

specific cases of vandalism even before those were reported to police, has made it the talking-point across UK and elsewhere. Significantly, the model could tell beforehand where the youths would likely to gather during the riots. The police, according to reports, is informed about the gathering only minutes after the model could predict. There were instances when the model could even predict the movement hours before the police came to know about those. 

“In this research we show that online social media are becoming the go-to place to report observations of everyday occurrences -- including social disorder and terrestrial criminal activity,” said a Cardiff University computer scientist, as reported in the University’s news release being published in various newspapers. 

2011 London riots, which were marked by large scale looting and vandalism, began at Tottenham after a young man was shot by police. 2011 riots was the worst riot in London in three decades.

The model developed by Cardiff University researchers can also find out whether riots are rumoured to take place or a group of youths is actually gathering at a particular area.

The University researchers have also tested the model after collecting as many as 1.6 million tweets relating to the 2011 London violence.