Link intelligence establishing neurocognitive knowledge-processing capabilities in a knowledge network

The objective of this paper is to demonstrate link intelligence that characterizes neurocognitive knowledge processing capabilities through dynamic knowledge development, learning and stability within a given knowledge network. The existing knowledge networks connect two or more data nodes with edges that have no intelligence embedded into them and provide for static information connectivity and retrieval. This paper focuses on the link intelligence that contributes towards the development of neurocognitive knowledge network model (NCKM) with autonomous processing nodes. Links in NCKM exhibit neurocognitive knowledge processing characteristics by virtue of its four properties viz. efficient knowledge assimilation, self-directivity, self-organization, and equilibrium. The simulation results comprise searching different concepts from NCKM to retrieve knowledge threads for the searched concept. The results exhibit the significance of link-weight gradation in self-directivity and self-organization of links for intelligent knowledge retrieval. The results also demonstrate the significance of equilibrium process to maintain stability in the knowledge network, by limiting the link-weight values within the saturation limits. Additionally, the results depict the criticality of coupling retrieval of random concepts with equilibrium process in providing knowledge build-up and learning within NCKM. NCKM with its intelligent links and processing nodes finds its applicability in many cognitive and intelligent systems.