Abstract: Artificial Intelligence is increasingly applied in domains where reliability, fairness, and transparency are essential. In this keynote, I will present recent contributions from my work on counterfactual explanations and fairness-aware preprocessing methods, two approaches that move beyond accuracy to ensure systems are both interpretable and equitable. Counterfactual explanations provide actionable insights, showing how specific changes in input can alter model outcomes, while fairness-oriented preprocessing addresses bias at the data level, reinforcing the central role of data-centric AI in building trustworthy systems.

To illustrate these ideas, I will draw on examples from clinical decision making, a representative high-risk scenario where explanations must be both understandable and actionable, and where fairness in patient data is critical. These case studies highlight the importance of designing AI systems under the principles of Trustworthy AI, while also considering the role of human-in-the-loop approaches as a way to align automation with expert oversight.

The objective of the talk is to show how these methods can translate theoretical advances into practice, bridging the gap between technical innovation and responsible deployment in sensitive domains such as healthcare.

Bio: Dr. Alberto Fernández Hilario (ORCID ID: 0000-0002-6480-8434) earned his Ph.D. in Computer Science from the University of Granada in 2010, supported by a prestigious FPI scholarship. Since February 2016, he has been affiliated with University of Granada, becoming Full Professor in the Department of Computer Science and Artificial Intelligence in 2022; he is also affiliated with the prolific Andalusian Data Science and Computational Intelligence Institute (DaSCI).

Dr. Fernández’s research spans Data Science, Big Data, Computational Intelligence, and Trustworthy AI, with particular focus on methods for explainability, fairness, and responsible model design. His influence in the field is reflected in over 25,000 citations and an h‑index of approximately 51 on Google Scholar.

He is the editor of Learning from Imbalanced Datasets (Springer, 2018), a seminal work that continues to attract significant scholarly attention (1,158 citations recorded on ResearchGate).

A highly successful mentor, Dr. Fernández has supervised five completed doctoral theses and currently oversees several additional Ph.D. projects in areas including complex Data Science problems, and both Trustworthy and ethical AI. He has extensive teaching experience: delivering over 250 ECTS in undergraduate courses and more than 15 ECTS in master’s programs, alongside supervising numerous Bachelor’s and Master’s theses.

He has also secured competitive research funding exceeding €500,000, notably including a Marie Skłodowska-Curie ITN H2020 project in bioinformatics and the participation in multiple competitive national projects and knowledge transfer contracts with important corporations.

His scientific excellence has been recognized with prestigious awards such as the Highly Cited Researcher in Computer Science (Clarivate, 2017) and inclusion among the top 2% most influential researchers globally (Stanford/Elsevier ranking).

Dr. Fernández’s research succeeds by combining theoretical rigor with societal relevance, promoting transparent, fair, and human-centered AI systems capable of real-world impact.