THE TASK APPROACH AND STRATEGIC SEMANTIC MODELING
V.S. Gumirov1, A.V. Gumirov2, D.I. Sviridenko3,2
1Independent researcher, Novosibirsk, Russia
2Novosibirsk National Research State University, Novosibirsk, Russia
3Institute of Philosophy and Law, Siberian Branch of the Russian Academy of Sciences, Novosibirsk, Russia
Keywords: computer science, mathematical logic, language, constructive model theory, semantics, methodology, programming theory, No-code and Low-code, executable specifications, problem, problem-solving criterion, problem-solving context, semantic modeling, ontology, artificial intelligence
Abstract
The article continues the discussion of the concept of semantic modeling developed by the authors. It examines the features, limitations, and characteristics of information systems created either using AI technologies or traditional methods. The possibility of expanding the provisions and tools of the semantic modeling concept is analyzed with regard to the situation of creating information systems focused on solving an open class of problems. It is proposed to develop behavioral strategies for such IT systems based on the possibility to leverage the experience of their previous actions in similar circumstances, selecting those actions that have previously yielded the best results, and then, using the task-based approach, select the most appropriate problem-solving methods under given conditions, either through simulation modeling or decision-making in order to determine the method most appropriate to the problem statement, paying particular attention to contextual conditions and the criterion for solution. The article presents and discusses the goal of further developing a semantic modeling methodology aimed at creating IT systems that behave in the manner described above, which explains the title of the article. In conclusion, it describes a possible variant of an extended semantic modeling language designed to specify the behavioral logic of such systems.
|
