Skip to main content

Artificial Intelligence and advanced Data Analytics for Law Enforcement Agencies

Periodic Reporting for period 1 - AIDA (Artificial Intelligence and advanced Data Analytics for Law Enforcement Agencies)

Reporting period: 2020-09-01 to 2021-08-31

In the framework of the "European Strategy on Artificial Intelligence", the role of AI as a key driver in solving some of the world's biggest challenges is clearly stated. The 'Communication on Artificial Intelligence' highlights the crucial role that AI technologies can play across the EU economy for a wide range of purposes, including supporting law enforcement (LEAs) in anticipating cyber security threats. Equipping LEAs with appropriate tools to ensure the safety of citizens while respecting their rights and freedoms, enabling them to detect fraud and cyber security threats and fight crime more efficiently, making the identification of online terrorist propaganda or the discovery of suspicious transactions more immediate, are only some of the opportunities that open up in front of a wise and appropriate use of AI technologies.
AIDA is positioned in this context, enabling collaborative services to help LEAs prevent, detect, counter and mitigate cybercrime and terrorism-related activities and events, including propaganda and radicalisation. Supporting LEAs to anticipate and respond more promptly to possible threats from cyber criminals and terrorists AIDA contributes to better protect EU citizens from crime and terrorist acts.
AIDA aims to provide EU LEAs with innovative, dedicated and sustainable digital technologies to facilitate the entire Big Data analytics cycle as part of their investigative and proactive prevention capabilities.
Overall objectives and key areas of innovation of AIDA are:
i/ Improving collaboration among EU LEAs and Europol
ii/ Data acquisition from open and privileged information sources
iii/ Meaningful processing and enrichment towards data that is ready for analysis
iv/ Analysis, reasoning, prediction and visualisation to obtain timely actionable intelligence
v/ Long-term strategic forecasting and analysis for the evolution of criminal threats and trends, and effectiveness of countermeasures
vi/ Societal acceptance and impact of such new capabilities on society’s sense of security and resilience
A key result of the project is a Security Sandbox allowing LEAs to review the new tools delivered, configure them according to their individual requirements and needs and to perform test sessions. Each tool will be autonomous, allowing easy integration and interoperability with pre-existent LEAs’ technological capability.
The overall implementation strategy adopted in the two-year project will focus on exploiting the full potential of Big Data by developing a set of AI-driven capabilities to maximise the Big Data acquisition, pre-processing, processing, enrichment and storage, and ultimately derive actionable intelligence on cybercrime and terrorism. The project adopts 10 work packages, specifically covering all the different aspects of the project.

WP1 coordinates all the activities within the project and monitors on its evolution ensuring the achievement of the project objectives within budget and scheduled deadlines while assuring the quality of project deliverables and the activities and results.
The aspects related to the conceptualization of the system ensuring that the user needs related to both security and interoperability are fulfilled are covered in WP2. The main results of the WP2 are therefore to define the use cases for Cybercrime and Counter-terrorism and the user scenarios guiding the development of the AIDA platform and related tools; to identify the data to be implemented in the scenarios and the AI functionalities to be included in the AIDA platform; to elicitate and formalise user requirements; to define system specifications and architecture.
In RP1, the main focus of WP3 was on covering the full range of possible societal and ethical concerns related to the interaction of LEAs with Big Data and the application of techniques in the field of ML and AI to policing paying special attention to cross- border exchange of information among law enforcement and data protection.
The analysis of the impact of increasing LEAs’ capabilities in the context of CC and CT on citizens’ security perceptions and of how technologies used in each project may impact their daily life is conducted in WP4. Preliminary results of WP4 concerns also the explanation of how such perceptions can be linked to citizens’ online and offline behaviours and are reported both in a project deliverable and in a paper.
In WP5, most of the work related to the implementation of the tools to find, acquire and manage relevant sources containing cybercrimes and terrorist-related activities and contents from the Surface, the Deep Web and the Darknet has been covered.
WP6 aims at delivering a set of smart multimodal processing services able to perform advanced analysis to extracting semantic information from the textual, audio and visual content.
WP7 provides capabilities for knowledge mining through services for discovering higher-level information, events, and facts by integrating and fusing the information extracted by the analysis tools coming from WP6. Visual analytics services will be provided for analysing in depth activities/events and monitoring dangerous situations, tracking terrorist group and identifying emerging trends.
WP8 will integrate and deploy into the overall system, the services resulting from WP5, 6, 7 and the applications for LEAs. WP deals with Pilots planning and preparation and system evaluation.
WP10 aims at planning, designing and implementing activities that will facilitate the achievement of the project’s impact. So far, activities have focused mainly on communicating the project to its intended audience and the general public, using a specific branding and messages and appropriate channels. More in details, two project’s newsletters have been issued, a printable flyer has been prepared and it is ready to be used by partners for any dissemination opportunity, an AIDA Community has been opened on Zenodo, ensuring open access to peer-reviewed publications stemming from AIDA and collecting all project publications on a single page, social media channels (LinkedIn and Twitter) have been launched. A number of relevant H2020 security funded projects, working towards similar goals as AIDA, have been identified.
AIDA will boost the effectiveness of the Security Union and mitigate / avoid potential consequences of crime-related incidents by improving the quality of data, and convert voluminous and heterogeneous data sets into actionable intelligence; Thanks to the use of domain-specific tools and Big Data analysis applications designed for the needs of crime investigators, AIDA will also contribute to a series of strategic, societal goals as well as to LEAs capabilities enhancing, and standardisation and policy impacts. Specifically,
Scientific and technical research impacts: pushing forwards the boundaries in research centring on Big Data analytics, ML and AI;
Socio-economic impact: thanks to the envisaged development and provision of innovations that meet European and global market needs, AIDA is expected to strengthen the competitiveness and growth of European companies;
AIDA Societal impacts: common European response to criminal and terrorist threats to citizens and society as a whole. Understanding of conditions for societal acceptance and effects on citizen behaviours & societal resilience.