Photo: Gabriela Pires
In his doctoral dissertation, MSc Tiago Brasileiro Araújo investigates how Entity Resolution systems, that is, AI-based technologies used to identify and merge records referring to the same real-world entity across different data sources, can be made fairer, more transparent, and better suited to continuously arriving data (streaming environments). Such systems are widely used in domains such as e-commerce, healthcare, digital services, and public administration, where inaccurate or biased decisions may affect individuals and organizations. The results demonstrate that fairness and explainability can be incorporated directly into the data matching process without substantially compromising accuracy or efficiency. The proposed approaches reduce bias in matching outcomes while providing understandable explanations for automated decisions, enabling more trustworthy and accountable AI systems. These findings are particularly relevant given the growing adoption of data matching technologies in critical digital infrastructures and decision-making processes. The dissertation provides valuable insights for researchers, practitioners, companies, and public organizations seeking to develop responsible AI solutions that are not only effective, but also transparent and equitable.
The doctoral dissertation of M.Sc. Tiago Brasileiro Araújo, in the field of Computer Science, titled Fairness- and Explainability-Aware Streaming Entity Resolution, will be publicly examined at the Faculty of Information Technology and Communication Sciences at Tampere University on 1 July 2026.
The Opponent will be Associate Professor Ester Zumpano from the University of Calabria, Italy. The Custos will be Professor Konstantinos Stefanidis from the Faculty of Information Technology and Communication Sciences, Tampere University.