Cognitive computing system based on distributed knowledge

Abstract

In the current paper the results of research and development of a computational system with cognitive properties based on inhomogeneous knowledge distributed by a set of Intelligent Agents are presented [1, 2, 4]. The Multi-Agent system represents a network with the Mesh topology in which each Agent , 1, i A i I is defined by: the initial state () i Xt, where N XR ; a strategy i S , or a set of data processing rules for system evolution; and a target () opt i XT, which also determines the purpose of the Multi-Agent system evolution. An Agent is defined as a computing structure with Wireless communication properties that manages two data structures: the knowledge storage model and the knowledge search model [3, 4]