Scouring the dark web for terrorist activity
Terrorists are increasingly exploiting technology in the age of the internet. The global network has become a powerful tool for terrorist organisations, allowing them to spread propaganda, recruit new members and fund and plan their activities. In recent years, terrorists have relocated to the Deep Web and Dark Nets, further obfuscating their presence from counter-terrorism forces. To help European law enforcement agencies combat this persistent threat to society, DANTE has created effective, efficient, automated data mining and analytics solutions and an integrated system to detect, retrieve, collect and analyse huge amount of heterogeneous and complex multimedia and multi-language terrorist-related information. The system works by ‘connecting the dots’ from discrete and seemingly unrelated online activity, to pinpoint potential terrorists. Seeking out the suspects Groups, leaders and suspicious actors can be detected, analysed and monitored at different levels, including sociological, criminological and psychological, to identify behavioural patterns. The DANTE framework and platform has the capability to gather, track and analyse information from messages, files and data exchanged online. It allows users to track people when acting on and moving from the public Web to the Dark Nets and vice versa, even if they change identities. “It is crucial to identify and recognise the real identity of people hidden behind the virtual accounts,” says Mr Ernesto La Mattina, Head of the Security Intelligence Unit at Engineering Ingegneria Informatica and DANTE project coordinator. The system automatically translates texts from several languages into English and summarises the information. Natural language processing algorithms process and extract relevant information. It can discover encoded messages and perform detailed linguistic analysis between and within texts. Multimedia contents also contain a lot of relevant information about activities and events. “One of the challenges of the project is to automatically identify and cluster/classify such activities and reconstruct the chain,” says La Mattina. With video and images, DANTE can pick out tampering and falsification, and detect and identify people, as well as designs such as logos and tattoos. Events and concepts can be recognised automatically, too. When analysing audio, DANTE can discern languages, translate from multiple languages straight into text format, and identify speakers and sounds. Money is the cornerstone of all terrorist activities; disrupting finances is a key element in fighting terrorism. Terrorists run online fundraising campaigns to support their activities and those of other illegal organisations. Another important fundraising channel is the illegal trafficking exploited through the internet – in particular selling illegal drugs and weapons in dark markets. “DANTE contributes to identifying terrorism fundraising campaigns on the internet through the detection of anomalous financial transactions and tries to connect the suspects with the digital (and afterwards the real) identity of authors,” says La Mattina. Spreading the word The DANTE platform has already been tested and used by the four law enforcement agencies involved in the project: ITCC (Carabinieri, Italy), PJ (Policia Judiciaria, Portugal), GUCI (Guardia Civil, Spain), and CAST (Home Office CAST, UK). Word of mouth has also led to overseas security outfits asking to trial the software. DANTE is widely known throughout the EU and around 3 000 officers have used the tools and services. Investigators who have tested the platform have expressed excellent feedback. “DANTE is for sure a success story,” says La Mattina.
DANTE, terrorist, finance, law enforcement, dark web, dark net, identities