DC3: Machine learning in digital finance

Objectives

The project aims to design and enhance the functionality and user interaction of applications that model Early Warning Signals. It will develop models and frameworks that leverage big data analytics, focusing on refining usability to improve clarity for various stakeholders. This project also seeks to implement and evaluate explainability methods to determine their impact on the decision-making processes of internal customers within financial institutions.

Expected Results

The project will develop and integrate advanced explainability features into various Early Warning Signal (EWS) applications, aiming to make analytical results and sentiment assessments more intuitive and actionable for users. Specifically, the project expects to (i) Enhance user understanding and interaction with big data outputs, promoting broader and more effective use across different business areas within financial institutions and regulatory bodies.

Secondments (3)

Recruitment & Hosting Details

DC 3 UTW University of Twente M9 36

4

DC 3 UTW UTW Month 9 36 months D 4.2 - 4.3

Machine learning in digital finance (WP 4)

Objectives: The project aims to design and enhance the functionality and user interaction of applications that model Early Warning Signals.

It will develop models and frameworks that leverage big data analytics, focusing on refining usability to improve clarity for various

stakeholders. This project also seeks to implement and evaluate explainability methods to determine their impact on the decision-making

processes of internal customers within financial institutions. The goal is to enhance their ability to provide accurate feedback and reporting

and to facilitate implementing necessary corrections.

Expected Results: The project will develop and integrate advanced explainability features into various Early Warning Signal (EWS)

applications, aiming to make analytical results and sentiment assessments more intuitive and actionable for users. Specifically, the project

expects to (i) Enhance user understanding and interaction with big data outputs, promoting broader and more effective use across different

business areas within financial institutions and regulatory bodies. This will involve improvements to make sentiment analysis results more

transparent, enabling supervisors to more effectively assess and verify these insights; (ii) Enables te assessment of the impact of these

enhancements on the decision-making processes of internal stakeholders, with the goal of establishing a feedback loop that encourages

continuous improvement of the platforms.

53 This candidate will be fully financed by the Swiss State Secretariat for Education, Research and Innovation (SERI). The SERI will apply the same rules as the European

Commission, based on the MSCA Work Programme 2021 - 2022, incl. recruiting, mobility and funding rules.

54 The University of Twente will be the formal degree awarding institute for the two DCs from BFH.

55 All months are in absolute terms since start of the project.

30

Project: 101119635 — DIGITAL — HORIZON-MSCA-DN-2022

Planned secondments: ECB, Lukasz Kubicki, M21, 12 months, exposure to globally leading central bank, research training on EU

principles, supervision

Fraunhofer, Prof. Dr. Ralf Korn, M33, 6 months, applied industry-research, contribute to multiple projects on blockchain and decentralized

finance

Swedbank, Prof. Dr. Tadas Gudaitis, M39, 6 months, ESG and credit score modelling

Fellow Host institution PhD enrolment Start date Duration Deliverables

4

Deliverables