Classification by ordinal sums of conjunctive and disjunctive functions for explainable AI and interpretable machine learning solutions

• Published in: Knowledge-based systems Vol. 220; p. 106916
• Main Authors: Hudec, Miroslav, Mináriková, Erika, Mesiar, Radko, Saranti, Anna, Holzinger, Andreas
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 23.05.2021; Elsevier Science Ltd
• Subjects: Agglomeration; Aggregation functions; Algorithms; Classification; Cognitive tasks; Domains; Explainable AI; Explainable artificial intelligence; Glass-box; Interactive ML; Interpretable Machine Learning (ML); Machine learning; Ordinal sums; Subject specialists; Sums; Transparency; Interpretable Machine Learning (ML); Transparency; Explainable AI; Interactive ML; Glass-box; Aggregation functions; Ordinal sums
• ISSN: 0950-7051; 1872-7409
• DOI: 10.1016/j.knosys.2021.106916

We propose a novel classification according to aggregation functions of mixed behaviour by variability in ordinal sums of conjunctive and disjunctive functions. Consequently, domain experts are empowered to assign only the most important observations regarding the considered attributes. This has the advantage that the variability of the functions provides opportunities for machine learning to learn the best possible option from the data. Moreover, such a solution is comprehensible, reproducible and explainable-per-design to domain experts. In this paper, we discuss the proposed approach with examples and outline the research steps in interactive machine learning with a human-in-the-loop over aggregation functions. Although human experts are not always able to explain anything either, they are sometimes able to bring in experience, contextual understanding and implicit knowledge, which is desirable in certain machine learning tasks and can contribute to the robustness of algorithms. The obtained theoretical results in ordinal sums are discussed and illustrated on examples. •Classification by neural networks is highly effective, but the solution is not re-traceable, hence interpretable.•Fuzzy classification is re-traceable, but a higher number of sets reduces interpretability.•A novel classification by the ordinal sums of conjunctive and disjunctive functions is proposed.