New Algorithms to Track Violent Groups on Twitter

This post is also available in: עברית (Hebrew)

New sentiment analysis algorithms are able to monitor the social network Twitter in search of violent groups. The system analyses both the messages they share and how their relationships develop.

Other uses of the system could be preventing phenomena such as bullying and racism. It can also be used to analyse what’s being said about companies, brands and people on social media.

According to Sinc, the tool was developed by researchers at the University of Salamanca, Spain (USAL). Spanish law enforcement authorities have already expressed interest in the tool.  “This system could have been very useful –for example– as a support system to control the violent football fans that caused serious incidents during Euro 2016 in France,” said Juan Manuel Corchado, professor of Computer Science and leader of the Bisite artificial intelligence team at USAL.

The police and other law enforcement agencies could use the tool to detect critical points, threats and areas with concentrations of potentially dangerous people. “It’s based both on the semantic analysis of messages and historical data and their evolution.”

The team used hybrid neural–symbolic artificial intelligence systems and methods that solve problems using a combination of different algorithms,

The new application can analyse emotions in different languages –at the moment, Spanish, English, French, German, Russian and Arabic. Also, it can extract communications information and see changes in individuals’ sentiments and physical location, analysing group interrelationships at the same time.

“It can establish where a dangerous user is located with reasonable precision, based on what they share on Twitter and how and with whom they are connecting at any time, without the need of geolocating tweets,” he adds.

Depending on the traffic generated between people, it is possible to identify the members of a group, its leaders and followers. It is also possible to see how relationships evolve and whether new members join a group. In addition, says the researcher, having identified the leaders, the tool can try to influence them to change their sentiments.

Spain’s law enforcement authorities will soon begin validation of the tool.

The tool will be put to market through partner companies of the Bisite group, which is very active in transferring technology to the corporate world.

The team is working with CSA company in a project called PIAR, which has enabled the development of highly sophisticated sentiment analysis algorithms that can be used in security applications.