A Neural Network Approach to Overcoming a Priori Uncertainty in Optimal Action Planning of Intelligent Information Agents for Soft Architectures of Service-oriented Systems

The necessity of developing soft architectures of service-oriented systems has been substantiated. The key properties of soft architectures of service-oriented systems are highlighted. Possession of key properties is associated with overcoming a priori uncertainty using neural network models. The functionality of soft architectures of service-oriented systems with intelligent information agents is described. The openness of the life cycle of soft service-oriented systems in relation to the methodology for designing action planners is shown. The connection between the methodology and the solution of the problem of choosing an algorithm for action planning is considered. The reasons for a priori uncertainty in knowledge about the analytical dependence of the planning efficiency on the conditions of the problem being solved and the characteristics of the subject area are revealed. A neural network approach to overcoming a priori uncertainty in knowledge with optimal action planning of intelligent information agents for soft architectures of service-oriented systems is proposed. The content of the end-to-end chain of stages in the implementation of the proposed approach is disclosed.