Mechanism of concept learning in topologized knowledge granular space based on GrC
Zuqiang Meng, Zhongzhi Shi, Shimo Shen · 2012
Concept learning in information systems is actually performed in knowledge granular space on information systems. But no much attention has been paid to study such a knowledge granular space and its structure so far, and its structure characteristics are still poorly understood. In this paper, the granular space is firstly topologized and is decomposed into granular worlds. Then it is modeled as a bounded lattice. Finally, by using graph theory, the bounded lattice obtained is expressed as a hass graph, and the mechanism of concept learning in information systems can be visually explained. With related properties of topological space, bounded lattice and graph theory, the "mysterious" granular space can be delved more deeply into. This work can form a basis for designing concept learning algorithm as well as can richen the theory system for granular computing.