Formulation of the problem and definition of approaches to building semantic knowledge models for artificial intelligence.

The article examines the issues related to the creation of semantic models of knowledge that can be used to endow artificial intelligence systems with the ability to understand the meaning of text in natural or any other language. Possible means for constructing semantic models of knowledge include the mechanism of multi-system integration of knowledge developed by the authors earlier, formal ontologies, and techniques of understanding meaning that have emerged within the framework of philological hermeneutics. Significant components of the presented study include an examination of the currently used language models of artificial intelligence, a new approach to the conceptualization of knowledge through its generalization in the form of open models, an assessment of the genesis and prospects of teleological and axiological interpretations of meaning for natural and artificial cognitive systems. The methodological basis of the presented study consists of the authors’ developments in the field of systems analysis, well-known analytical methods adopted within hermeneutics, structuralism, classical epistemology, formal ontology theory, and linguistic and language modeling. The scientific novelty of this research lies in the determination of the necessary tools for creating semantic models that generalize knowledge. The mentioned tools include: multi-system integration of knowledge based on the integration of the subject of cognition into multiple systems with subsequent generalization of the patterns identified in these systems and their translation for solving tasks of understanding and creativity; formal ontologies that implement the description of knowledge from a specific domain in the form of conceptual schemes, taking into account existing rules and relationships between elements, allowing automatic extraction of knowledge; and a wide variety of hermeneutic techniques for understanding meanings. Objective limitations of use for artificial cognitive systems that lack subjectivity and value prioritization in understanding meanings are noted. Some limitations in the use for artificial cognitive systems are also found in hermeneutic techniques for understanding the meaning of text. This is related to the impossibility of full reflection without feelings, emotions, and desires generated by needs that also initiate subjectivity.

semantic model, knowledge, artificial intelligence, cognitive system, multi-system integration of knowledge, patterns, formal ontologies, value prioritization, needs, subjectivity