Skip to main content
Go to the home page of the European Commission (opens in new window)
English English
CORDIS - EU research results
CORDIS

Comprehensive data-driven Risk and Threat Assessment Methods for the Early and Reliable Identification, Validation and Analysis of migration-related risks

Periodic Reporting for period 2 - CRiTERIA (Comprehensive data-driven Risk and Threat Assessment Methods for the Early and Reliable Identification, Validation and Analysis of migration-related risks)

Reporting period: 2023-03-01 to 2024-08-31

Migration implies several societal opportunities and challenges. The migration process itself is coupled with a number of interrelated risks and threats -such as human trafficking and public health threats - for all stakeholders including border agencies, NGOs and the migrants themselves. This situation implies a need for more cross-sectoral, transdisciplinary and cross-country cooperation in all risk management phases. Existing risk analysis models have a number of shortcomings such as limited consideration of humanitarian aspects, which prevent the unfolding of their full potential.

The CRiTERIA project seeks to strengthen and expand existing risk analysis methods, by introducing novel approaches, such as identifying risk factors from qualitative evidence, building composite indicators, incorporating risk interaction and risk cascading assessments and consolidating the human security and human rights dimensions of border security. Strong and accurate risk and vulnerability analysis models have to be backed by effective intelligent analysis technology and tools. In CRiTERIA, we will develop and evaluate advanced analysis technologies and tools that are tailored to the new comprehensive risk and vulnerability indicators of the methodology.

In summary, the goal of the CRiTERIA project is a novel, comprehensive but feasible and human-rights sensitive risk and vulnerability analysis framework for border agencies and other stakeholders, which backs a novel multi-perspective risk and vulnerability analysis methodology with multi-source, multi-lingual analysis technologies and tools for serving the complex indicators of the methodology and for making them accessible in a verifiable and understandable way. The methodology will be developed and validated in close collaboration with practitioners from border agencies.

For achieving its goal, five objectives have been defined: 1) a refined multi-perspective risk, vulnerability and threat understanding; 2) tools for effective cross-media and multi-source risk and vulnerability analysis including validation support; 3) awareness for the evolutionary character of the realm of observation; 4) integrative and impactful solutions, which can be easily adopted; 5) actionable insights, which help decision makers and border agencies to counteract risks as well as the negative impacts resulting from them; and 6) legally compliant and societally acceptable solutions.
The CRiTERIA project has achieved its goal. Major effort has been invested in better understanding the risk analysis practices of CRiTERIA end users and external stakeholders. As a basis for the improved risk analysis methods, a mapping of sources has been created for relevant open source information. The collection and organization of risk factors for the risk analysis method was driven by a set of jointly identified project use cases. Furthermore, a methodology for the composite indicator for vulnerability of migrants (Human Security Filter) was developed. This indicator will complement the existing CIRAM risk analysis framework by putting a larger emphasis on the wellbeing of migrants and human security aspects.

Major progress has also been made in technology development for effective data and cross-media analysis in support of risk analysis. Various components have been developed including (improved) methods for free-text search in large video collections, for logo detection, for the detection and analysis of narratives and for the automated detection of risk factors and for automated speech recognition for a wide range of languages. For explainable evidence validation, which provides the basis for decision support in risk analysis, a variety of methods have been developed. This includes image manipulation detection and an interpretable open-domain fact-checking model.

Several releases of the CRiTERIA platform have been developed, which is equipped with two powerful UIs (tools): the CRiTERIA Dasboard, which supports the exploration phase in risk analysis and the risk analysis tool CREST for the risk analsis and validation phase. The project results have been validated in an intensive piloting activity in cooperation with the LEA partners in the project.

The consideration of legal, ethical and societal aspects was also an important part of the project work. Here, a human rights analysis has been performed resulting in a human rights checklist. In the second phase of the project, an intensive analysis of current legal frameworks such as the AI Act and the EU Pact on Migration and Asylum have been performed and policy recommendations have been derived from these analyses. For addressing societal issues, an intensive analysis has been performed with a focus on societal acceptance and critical factors as well as controversial topics for the acceptance of AI-based systems in the context of migration have been identified.

Starting early in the project, a common communication, dissemination and exploitation strategy was developed and activities started including the preparation of a number of events and training activities. In addition, a large number scientific publications have been achieved based on CRiTERIA results and project results have been presented at different types of events. The exploitation strategy has been refined and elaborated in the final phase of the project with a special focus on component-wise exploitation.
Progress beyond the state of the art has been made in the fields of visual media analysis and annotation, trend detection and prediction, visual media verification, detection of narratives, information extraction and linking, automatic speech recognition, risk analysis as well as in the field of privacy and data protection.

The project has created outcomes on different levels: a) methodology for comprehensive and multi-perspective risk and vulnerability analysis; b) innovative and effective IT and AI methods in support of the CRiTERIA approach for risk, vulnerability and threat analysis; and c) the legally compliant and societally acceptable CRiTERIA system, a flexible and extensible platform and tools. Furthermore, best practices, actionable insights and policy recommendation have been made available. Results of the project are made available in an easy to use form for multiple target groups in the CRiTERIA Toolkit.

The results of CRiTERIA are expected to create in impact in relevant areas: They will provide knowledge and evidence-based support to policy developments and will help to better manage the complexity of migration-related issues and the resulting risks and vulnerability. The design of the project ensures that both a) a comprehensive picture of the complexities is taken and b) the information is filtered and aggregated to help in managing the complexities and to create actionable insights. The development and use of composite indicators will ensure that the threats identified by the system can be better categorised and prioritised.

As a broader socio-economic impact, CRiTERIA fosters a risk understanding (based on the multi-disciplinary approach) that takes into consideration all stakeholders at risk in the migration process including the migrants themselves. This can change the understanding of the migration process, related vulnerabilities and required priorities in the mind set of border agencies, policy makers and the general public.
Method-Technology Interaction in the CRiTERIA Project
My booklet 0 0