Automation system of intellectual activity on creating programs in the language of logical programming

Abstract

The article presents a developed software package that implements the training system that automates intellectual activity in creating programs in the language of logical programming. The choice of semantic networks as a way of the representation of knowledge in the process of modeling subject domains in the learning system during the formulation of educational logical problems is substantiated. The typology of the learning tasks used to master the logical programming language Prolog is grounded, including the following types of tasks: logical, arithmetic, value analysis tasks with subsequent selection or ordering and creation of knowledge bases on the subject domain. In this case, for each type of problem, the possibility of applying analysis of formal concepts to solve them has been investigated and a set of examples has been compiled. The technique of analysis of formal concepts and their grouping depending on the parameters of objects of the domain is developed, the essence of which is to identify an invariant set of descriptors suitable for representing different subject areas, which allows the generation of rules. The technique for generating domain representation rules based on the declaration of predicates with one or more parameters, which are correlated with the peculiarities of formulations of typical learning tasks, is developed.