Zengenti.Contensis.Delivery.Entry Modelling migrant flows

Modelling migrant flows

AI proven unreliable and unsuitable, new expert systems are the way forward

The EU and states have been warned that AI is dangerous and unreliable for migration flow predictions thanks to a €4.87m research project to predict and manage migrant flows.

Brunel University London has been one of 14 institutions across Europe set to work on ITFLOWS, a 3-year project led by the Autonomous University of Barcelona and funded by the European Commission's Horizon 2020 research and innovation programme.

Brunel Law School and the Department of Computer Science were to use €595,500 share of 4.87million Euros to develop the evidence-based EUMigraTool, models of migrant flows and a report on the legal frameworks needed to offer the right services. However, the work found that AI models are not the way forward and expert systems offer better solutions for migration predictions, The EUMigraTool was a tool that would be created to provide improved predictions of migration and management of asylum seekers and refugees in the EU through the identification of hubs of tensions between migrants and EU citizens. Instead, prof Alexandra Xanthaki and Dr Ermioni Xanthopoulou proved that such an AI tool would be unreliable, as the data fed to it was not coherent, and dangerous, as it could be used to restrict migrant entry rather than accommodate migrants who come to the EU region. These results were shared with policy makers, civil society and scholarship.

On this basis, EUMigraTool was abandoned and its main non-AI tool, developed by our Brunel team led by Pro Derek Groen, evolved into a foresight approach that went incorporated expert knowledge rather than personal data. The vision to shift from forecasting to foresight was also driven strongly by UNHCR. This led to a foresight analytics tool developed by Brunel for the UNHCR to understand how many Ukrainians may be returning. In its further developed form, such a non AI tool has been now been used in six UNHCR projects. 

More work continues to be done both on the dangers of AI in migration and the new non-AI tool that would facilitate migration predictions for human rights guarantees.

View on YouTube

Please accept all cookies to view this content. Alternatively view the content on YouTube

Graphical abstract for the EUMigraTool
Graphical abstract for the EUMigraTool

The EUMigraTool will include analysis of media content from tv-news (video content), web-news and social media (text content) using deep learning and proposing novel deep architectures in generative modeling and forecasting using sequential data. 

The EUMigraTool

The EUMigraTool includes a hybrid model that combines machine learning with agent-based modelling. These models and predictions will support strategic decisions particularly for crisis situations, and help with the design and implementation of operational activities in that context.

Migrant flow predictions
Migrant flow predictions

Meet the Principal Investigator(s) for the project

Professor Alexandra Xanthaki
Professor Alexandra Xanthaki - Alexandra is a leading expert on indigenous rights in international law. AMong her several publications, her monograph Indigenous Rights and United Nations Standards: Self-determination, Culture and Land (Cambridge University Press) is considered a reference source on the topic. In 2011 Alexandra co-edited Reflections on the Declaration on the Rights of Indigenous Peoples (Hart) and most recently, in 2017, Indigenous Peoples' Cultural Heritage (Martinus Nijhoff/ Brill). Her work has been cited repeatedly in United Nations documents and she has given keynote speeches around the world, including the Arctic Centre, Rovaniemi; the KL Bar, Malaysia; Trento, Italy; and London. She has worked closely with the UN Expert Mechanism on the Rights of Indigenous Peoples, the UN Special Rapporteur on Indigenous Issues, the ILO. Currently she is working with Minority Rights Group International on the rights of the Latin American community in the 7sisters re-development in Haringey, London. She has taught civil servants, indigenous leaders and activities in Vietnam, Pretoria, Kyiv, and London. She is a Senior Research Fellow at the Institute of Advanced Legal Studies (University of London).   Before she joined Brunel university, Alexandra taught in Keele and Liverpool. She has received the STAR award for her teaching and stduent support. She is a member of the Human Rights Faculty of the Centre for Continuing Education at the University of Oxford and has been an external examiner in several law departments, currently at Birkbeck.   Since October 2015, Alexandra leads the Athens Refugee Project, where she takes Brunel law students to Athens to volunteer in migrant and refugee sites, provide assistance and learn more on the refugee crisis in Europe from discussions with state authoriites, NGOs and IGOs. She has found invaluable partners in Maria Voutsinou from the Greek Ombudsman for Human Rights and Kenneth Hansen from Faros ('The Lighthouse'), an NGO on unaccompanied minors. Brunel University has received a congratulatory letter from the Greek state for this project. In 2017, Alexandra organised a series of academic multi-disciplinary events on Migrant and Refugee Rights in London (with IALS) and Athens.    Qualifications: LLB (Athens); LLM (QUB); PhD (Keele); Lawyer (Athens Bar)

Related Research Group(s)

kids on laptop

Computer Science for Social Good - Our group works with partners in the Global South to lead and promote interdisciplinary research in the field of computer science and social good. We focus on investigating and developing new ways and innovative technologies to address challenging socio-economic problems.


Partnering with confidence

Organisations interested in our research can partner with us with confidence backed by an external and independent benchmark: The Knowledge Exchange Framework. Read more.


Project last modified 05/11/2026