PRESERVE aims to improve the daily investigative and preventive work of Law Enforcement Authorities (LEAs), enhance proactive threat detection and response, and provide a regulatory-compliant framework for the responsible use of AI in crime detection and prevention.
The project will deliver a significant advancement in how LEAs can use technology to improve cross-border collaboration and decision-making while adhering to the ethical and legal standards of the European Union.
The €6 million project, funded by the European Commission’s Horizon Europe programme, involves partners from 8 countries across Europe – including researchers, law enforcement agencies from Greece, Spain, Romania and Czechia, and industrial partners with expertise in technology and software development.
Over the next 36 months, the project will develop a comprehensive solution to tackle critical challenges in public safety by efficiently and securely collecting and processing large amounts of data from various sources and performing trend analysis and real-time monitoring.
The project will focus on preventing child sexual abuse and facilitating the identification and mitigation of hate speech through trend analysis and real-time monitoring. It will analyse radicalisation risks and assess extremist activity using advanced tools, and will examine methods for detecting drug trafficking through advanced data correlation and trend analysis.
Professor Hamid Bouchachia
Through PRESERVE we are investigating ways to fight various types of crime that are instigated and amplified in cyberspace...resulting in ethical AI and machine learning-based technology that assists in combating cybercrimes.
PRESERVE intends to design, implement, and validate an advanced set of tools that enable security authorities to collaborate more effectively and securely without compromising citizens’ privacy.
It will use federated learning technologies to enable AI models to be trained in a decentralized manner, avoiding sharing of data to be shared among countries.
Bournemouth University, under the lead of Professor Hamid Bouchachia, will contribute to the development of such digital tools based on AI and machine learning.
Specifically, BU will contribute to a number of aspects:
- Federated learning to cope with multi-tenant data owned by LEAs.
- AI algorithms for online behaviour and cybercrime detection.
- Natural language understanding, video and image processing for cybercrime detection and investigation.
- Fair, ethical and transparent machine learning.