Students and professors of the Azerbaijan State Oil and Industry University (ASOIU) have achieved a remarkable milestone in the field of artificial intelligence. The article, co-authored by Pavel Kosov (a student at the ASOIU), Dr. Nahla El Kadhi and Professor Cecilia Zanni-Merk (UFAZ faculty), and Professor Latafat Gardashova (research professor at ASOIU), was published in the “Procedia Computer Science” journal by Elsevier, which is indexed in leading international scientific databases and ranked in Q2 according to the Scopus system.
The article titled “Formalizing fuzzy explainability: Enhancing explainable artificial intelligence with fuzzy-ontology-based explanatory properties” was published as part of the proceedings of the 29th International Conference on Knowledge-Based and Intelligent Information & Engineering Systems (KES-2025) in Volume 270 of the journal, on pages 3768–3777.
This research is considered one of the first steps in the field of explainable artificial intelligence (XAI) in Azerbaijan. In the paper, the authors build upon their previous work by combining previously developed explanatory properties with fuzzy logic to formally define the concept of “fuzzy explainability.” The proposed approach addresses the analysis of uncertain data and shared knowledge, which are common in real-world contexts. The framework is built on fuzzy OWL2 ontology, where the fuzzy membership degrees serve as a measure of certainty, thus making AI-generated explanations more trustworthy and human-understandable.
The researchers also expressed their gratitude to the organizers of the KES-2025 international conference and emphasized their intention to continue expanding their work in the critical area of AI explainability.
The full text of the article is available at the following link:
