Fuzzy Modeling, Granular Computing, and XAI
XAI enables AI systems to explain their decisions, often in natural language, and can benefit from the methodologies that Granular Computing offers for managing imprecision and uncertainty. As a cornerstone of Granular Computing, Fuzzy Logic provides the mathematical tools to develop interpretable models that can process graded concepts, define the meaning of most natural language terms and handle the uncertainty that characterizes decision support tasks. We use fuzzy models in several domains, including medical diagnosis and learning analytics.
Some recent publications
Kaczmarek-Majer, K., Casalino, G., Castellano, G., Dominiak, M., Hryniewicz, O., Kamińska, O., ... & Díaz-Rodríguez, N. (2022). PLENARY: Explaining black-box models in natural language through fuzzy linguistic summaries. Information Sciences.
Kaczmarek-Majer, K., Casalino, G., Castellano, G., Hryniewicz, O., & Dominiak, M. (2022). Explaining smartphone-based acoustic data in bipolar disorder: Semi-supervised fuzzy clustering and relative linguistic summaries. Information Sciences.
Casalino, G., Castellano, G., Castiello, C., & Mencar, C. (2022). Effect of fuzziness in fuzzy rule-based classifiers defined by strong fuzzy partitions and winner-takes-all inference. Soft Computing.
Moral, A., Castiello, C., Magdalena, L., & Mencar, C. (2021). Explainable fuzzy systems. Springer International Publishing.