Abstract
Humans' representation in nature language about geographic phenomena is usually qualitative rather than quantitative. Qualitative spatial reasoning provides an approach which is considered to be closer to the representation. Commercial GIS software are confronted with a challenge that the software should be equipped with artificial intelligent functions like qualitative spatial reasoning for more and more users, especially for spatial decision-makers. This paper proposes a framework of field-based fuzzy spatial reasoning through which qualitative description usually encountered in spatial reasoning process can be handled quantitatively. As preconditioning, field-based fuzzy representation structure for qualitative description is put forward, then the methods of constructing membership function are discussed. Standard operations of field-based fuzzy spatial reasoning model in the case of constraint satisfaction problem (CSP) are illustrated. An example explains the implication of the model in spatial decision-making process.
