The paper presents a model of comprehensive support for the development of intelligent DSS. The model includes the concept of comprehensive support that describes the needs of developers, methods and means for supporting the development of systems of this class, the structure of the repository of decision-making methods, the architecture of a typical DSS and the methodology for their development. The methodology is based on such principles as the maximum use of ready-made solutions, scalability, accessibility, openness, usability, independence from subject domain, and selfdescriptiveness. These principles have proven themselves in practice. Abidance of these principles is ensured by the use of ontological, fractal-stratified, service-oriented and wire-frame approaches, as well as rapid prototyping and flexible development approaches. The methodology involves the use of Semantic Web technology and technology for the development of intelligent scientific Internet resources. Such means of comprehensive support for the development of intelligent DSS as the ontology of the knowledge area "Decision-making Support", the information and analytical Internet resource, the repository of decision-making support methods, and the methodology for developing intelligent DSS are described in detail. It is proposed to use information and analytical Internet resource, which is built on the basis of the ontology of the subject area and the shell proposed by the technology as a framework of future intelligent DSS. The functionality of the DSS under development is provided by using services that are connected to the resource and implement methods stored in the repository. The language of description logics of the SOIN (D) family is used to describe the model of comprehensive support.

No abstract

No abstract

Computational plasma physics is a wide area of research, which includes a certain set of physical phenomena, mathematical equations, numerical methods, programming strategies, and hardware architectures that directly follow each other. To design an efficient code for solving a computational plasma physics problem, the researcher should have a clear understanding of the relationships between the concepts of physics, mathematics, computer science, and computer architecture. Ontology is one of the effective ways to realize (provide) such an understanding. Authors propose an approach to design an ontology which will show how physical phenomena determine mathematical equations being used, how the equations define numerical methods, and how methods enable programming strategies to form an architecture-efficient implementation of problem solution.

No abstract