Golian N. Predicative model of implicit relations in process knowledge representation

Object of the research is the processes of knowledge representation in artificial intelligence systems. The aim of the research - developing models of implicit relations in representing the process about knowledge to allow an informed choice of an action-based process for transforming implicit knowledge into explicit oneof the process of knowledge representation on the basis of analysis of the relevant sets of events. Research methods - methods of logical analysis, algebra, finite predicates and predicate operations for the construction of relational networks and predicate models implicit connections in the process representation of knowledge, methods of presentation and manipulation of knowledge in artificial intelligence systems and methods of mining processes for modeling of the process of knowledge representation. Equipment - personal computer. Theoretical and practical results of researches - development of models of presentation of hidden dependencies in the process representation of knowledge in order to provide the opportunity to make informed choices of action of the process on the basis of a transformation of hidden knowledge to explicit; model brought to the implementation of the programme, which allowed to construct relational network search implicit connections knowledge base for implementation in cerebra ling similar computers. Scientific novelty -for the first time the algebro-logic model of the generalized construction of an implicit choice in process knowledge representation, based on the model of an implicit connection between the events. This gives the opportunity to make clear the hidden choice between events and improve the accuracy of the model; for the first time offered predicate model implicit connections for input and output between the process and the external подпроцессом by process knowledge representation, which connect the parallel fragments of knowledge representation and are characterized by a set of necessary and sufficient conditions for the existence of parallel connections. Models provide the possibility of their detection and elimination of contradictions in the process representation of knowledge; improved predicate model of indirect relations between the events, which represent the steps in a process knowledge representation. Model, unlike the existing ones, takes into account consistently the occurrence of events and the absence of cycles of events that increases the accuracy and adequacy of the model; further development of the model of typical situations implicit choice in process knowledge representation in the form of relational networks, which allowed to find implicit links by comparing the States of the relational network to the sequence of the events of the process. The results of dissertation researchers have found practical application in Public Company Kharkiv machine-building plant "Svitlo shakhtarya" to improve the quality of information processing in problems of modeling of business-processes; reduce the amount of manual work and the time of formation of packages of training materials for computerized training systems and retraining of the personnel of the organization. The theoretical results of the thesis have been used in the educational process at the departments of the software engineering and applied mathematics of the Kharkov national University of radio electronics in the preparation of lectures on courses "Theory of intelligence" and "Algebraic logic" for a speciality "Software engineering". Scientific and practical results of the thesis can be used in systems of artificial intellect in finding implicit dependencies for processes of different nature (informational, social, economic); when modeling the processing of hidden knowledge of the natural intelligence; in the educational process in the preparation of specialists in the field of software engineering.