Periodic Reporting for period 2 - RED-Alert (Real-time Early Detection and Alert System for Online Terrorist Content based on Natural Language Processing, Social Network Analysis, Artificial Intelligence and Complex Event Processing)
Okres sprawozdawczy: 2018-12-01 do 2020-09-30
The research addresses threats to society (terrorism) and therefore will have specific societal impact in line with the H2020 “Security Societies” challenge.
Overall objectives of this project are the following:
1.Provide a complete toolkit for LEAs to collect, process, visualize and store online data related to terrorist groups, whether related to propaganda, fundraising, recruitment and mobilization, networking, information sharing, planning/coordination, data manipulation and misinformation.
2.Cover a wide range of social media channels, in particular new targeted channels, which are increasingly used by terrorist groups to disseminate their content.
3.Allow LEAs to take coordinated action in real-time while preserving the privacy of citizens.
Achievements:
- System architecture consolidated based on LEAs requirements and uses cases produced during the first months of the project. This result belongs to Work Package 1: "System Specifications & Architecture" from Description of Action;
- Core modules of the RED-Alert solution ready (WP2, WP3,WP4,WP5).This result is associated with the work performed according to objectives from:
Work Package 2: "Social Language Processing"
Work Package 3: "Social Network Analysis"
Work Package 4: "Complex Event Processing"
Work Package 5: "Privacy, Visualization and Meta-Learning"
from Description of Action;
- First release for the integrated RED-Alert solution completed. This result is associated with the work performed according to objectives from: Work Package 6: "Solution Integration";
- Ethics compliance ensured by elaboration of the procedures/policy to be covered by the consortium for different phases of the project (WP11). This result is associated with the work performed according to objectives from: Work Package 11: "Ethics requirements";
- Dissemination and exploitation plans elaborated and partially implemented. result is associated with the work performed according to objectives from:
Work Package 8: "Dissemination & Ecosystem Development"
Work Package 9: "Exploitation & Upscaling"
- Consortium cohesion and collaboration enhanced and improved under sustained effort from Work Package 10: "Management".
The development efforts are driven by end-user (LEA) requirements in order to make sure that they are aligned with real-world requirements. End-users selected for the project cover a variety of countries where the terrorism pressure is medium (Spain, Romania, Moldavia) to very high (UK, Israel) so the system can be field-tested in the most demanding contexts.
The project will therefore have a direct impact on the growth of the industrial suppliers of the consortium:
• INSKT, INT, MAV, ICE will leverage the project to develop their sales of customized social medial analytic products towards European LEAs and even more at international level;
• INSKT and INT will also be able to reuse the project know-how to enhance their existing NLP technology platforms (Sento and Intuscan) in various application domains (non LEA related) such as consumer sentiment analysis.
• SIV will use the know how gained to open new markets and to identify new opportunities for the current clients and for potential clients.
The technical work of the project requires an interdisciplinary approach of the project (social network analysis, natural language processing, semantic media analysis, complex event processing, data privacy, psychology, counter-terrorism) that can only be provided by a combination of researchers, industry suppliers and end-users.
The RED-Alert project consortium was developed in order to balance the contribution of:
• Academic researchers;
• Industry-driven innovation partners;
• End-users driven innovation.
New knowledge on NLP, SNA or CEP created by the RED-Alert research is the source of improvement for industry suppliers and in return, new market prospects for innovation identified by LEAs can point towards new avenues for intelligent law enforcement solutions.
Societal impact: giving better tools to European LEAs while strengthening personal data privacy. The project results will help LEAs in Europe (and around the world) to embrace digital technology and use social media to support their day-to-day activities.
As dependence on social media for personal and business information gathering and dissemination grows, social media use by law enforcement agencies will accelerate rapidly, prompting the need for social media training, policies, dedicated staff, and investment in social media analytics software, all of which can be supported by the RED-Alert project results. The project results will also help LEAs using social media must keep on top of legal and regulatory compliance issues, such as public records, data security, and terms of use with social media providers, jurisdictional considerations, and social media as it related to existing laws, such as those around harassment. Any project which involves the processing of personal data needs to strike a balance between the interests of society and the fundamental privacy rights of internet users, and this project will comply with generic data privacy legislation as well as specific substantive provisions relating to the processing of personal data within a law enforcement context.
The expected impact:
Impact 1:
• Improved investigation capabilities;
• Crimes solved more rapidly, to reduce societal distress, investigative costs and the impact on victims and their relatives;
• Prevention of more terrorist endeavors;
• Better identification and understanding of criminal activities.
Impact 2:
• LEA officers provided with better tools to help them on their (specialized) daily work.
Other impacts expected related to these project objectives are the following:
• Effective prevention of terrorist activities organized via the Internet;
• Increased competitiveness of suppliers.