Designing a knowledge base for the development of intelligent models based on the conceptual structure of activity act
Autor: | E. V. Brazhnikova, A. B. Sorokin, L. M. Zheleznyak |
---|---|
Rok vydání: | 2020 |
Předmět: | |
Zdroj: | Journal of Physics: Conference Series. 1615:012023 |
ISSN: | 1742-6596 1742-6588 |
Popis: | This article discusses the development of intelligent models based on the conceptual understanding of a dynamically complex environment. The method proposed in this work allows solving these problems based on a synthesis of the activity approach and situational analysis. As a result, the conceptual structure of the act of activity is realized. The combination of these structures leads to the construction of a decision matrix, which represents a knowledge base about a dynamically complex environment. An analysis of these interrelated aspects leads to the construction of four conceptual plans: the functional structure, processes, context and regularity. Conceptual plans make it possible to combine model ideas about the subject area, thereby simplifying the choice of intelligent modules for decision support systems. Recognition of the task of constructing conceptual structures as non-trivial, complex and time-consuming led to the idea of implementing the software system “Designer + Solver + Interpreter”. The software toolkit allows not only visualization of conceptual structures and implementation of knowledge bases for intelligent models, but also the studies for completeness and adequacy. |
Databáze: | OpenAIRE |
Externí odkaz: |