The Possibility of Applying the Theory of Multioperations to Build Self-learning Artificial Intelligence Systems with the Explainability Property

When developing artificial intelligence systems, one of the important aspects is a more accurate description of the considered events and their uncertainties. At the moment, most of the modern solution's mathematical apparatus have a significant drawback – the data is considered only between two logical elements – true and false. In this regard, it is impossible to give detailed description of the considered events and their uncertainties. Another important aspect of Artificial Intelligence systems is the explainability property, since making key decisions, it is necessary to explain of how the AI system came to its answer. One of the possible solutions is to apply the theory of multioperations. The paper suggests a methodology for building self-learning systems of explainable artificial intelligence based on the theory of multioperations and presents an abstract example of the application of the developed methodology for building an AI system of a medical field of study.