Implementation of scalable intelligent systems and applications based on mobile agents

Abstract

Background. A method for implementing a distributed metacomputer application in a computer network environment is proposed and tested in practice, based on the integrated use of a domain-specific multi-agent system and the concept of agent-oriented programming in accordance with given conceptual graphs. The purpose of the study is to develop recommendations on the use of logical methods and models of artificial intelligence when creating scalable intelligent applications. The solution of this problem is relevant and is associated with the extraction of structured information from large volumes of data and its subsequent processing for placement in databases and knowledge. Materials and methods. Conceptual and other models of artificial intelligence were used to ensure the interoperability of intelligent systems at the levels of external and internal design, i.e. mutual consistency of the tasks to be solved with the design methods of the intelligent system itself. Results. The technology allows scaling computing resources by adding new agents to a group, instances of which perform the specified operations in parallel and independently of each other. Implemented a dynamic search for agents for their interaction “hurriedly”. Conclusions. A method for implementing a distributed metacomputer application in a computer network environment is proposed and tested in practice, it is based on the integrated use of a domain-specific multi-agent system and the concept of agent-oriented programming in accordance with given conceptual graphs. Unlike other methods, it is proposed to create the program code directly according to the conceptual graph. An open, dynamically developing agent platform is used, which ensures the coordinated work of agents in accordance with the relations of the conceptual graph. The proposed technique allows you to quickly create distributed scalable applications for processing big data.

Keywords

• scalable computing
• mobile agents
• distributed computing
• logical model
• conceptual model
• system and functional architecture interoperability