To fight the war against terror, Law Enforcement Agencies (LEAs) are increasingly relying on social media intelligence (SOCMINT), a new field of intelligence covering a wide range of applications, techniques and capabilities analysing social media data, such as Natural Language Processing (NLP), Social Network Analysis (SNA), Artificial Intelligence (AI) and Complex Event Processing (CEP).
Commercial tools used by LEAs focus primarily on data mining and predictive analytics i.e. visualizing and analysing data on maps through time-sequence playback, centralizing multiple data stores in one system, discovering hidden value in existing information stores, sharing data with other law enforcement organizations.
The RED-Alert solution will cover a wide range of social media channels, in particular new channels such as Telegram and Periscope, which are increasingly used by terrorist groups to disseminate their content. The RED-Alert solution will allow LEAs to take coordinated action in real-time while preserving the privacy of citizens.
The specific objectives of RED-Alert are:
Objective 1: Improve available building blocks (NLP, SNA, CEP) and combine them with artificial intelligence (AI) in order to support the language, usability and privacy-preserving requirements of the RED-Alert solution
Objective 2: Integrate all building blocks (NLP, SNA, CEP) in a holistic solution with real-time collaborative capabilities and create a baseline system (development environment) for current and future use
Objective 3: Involve law enforcement agencies in the specification and the demonstration of the RED-Alert solution
Objective 4: Disseminate the project results in the scientific and law enforcement communities, prepare the exploitation of the RED-Alert solution and its individual modules
RED-Alert will outperform state-of-the-art solutions in terms of performance (number of languages supported, number of social media channels mined, accuracy), usability (visualization, real-time operation, collaboration, development support) and privacy protection.