Method and application of comprehensive knowledge discovery
Tóm tắt
This paper proposes the principle of comprehensive knowledge discovery. Unlike most of the current knowledge discovery methods, the comprehensive knowledge discovery considers both the spatial relations and attributes of spatial entities or objects. We introduce the theory of spatial knowledge expression system and some concepts including comprehensive knowledge discovery and spatial union information table (SUIT). In theory, SUIT records all information contained in the studied objects, but in reality, because of the complexity and varieties of spatial relations, only those factors of interest to us are selected. In order to find out the comprehensive knowledge from spatial databases, an efficient comprehensive knowledge discovery algorithm called recycled algorithm (RAR) is suggested.
Tài liệu tham khảo
Yee L, Ma J H, Zhang W X (2001) New method for mining regression classes in large data sets.IEEE Transactions on Pattern Analysis and Machine Intelligence, 23(1): 5–21
Lu W, Han J, Oci B C. (1993) Discovery of general knowledge in large spatial databases. Far East Workshop on Geographic information Systems, Singapore.
Taylor K, Walker G, Abel D (1999) A framework for model integration in spatial decision support systems.Int. J. Geographical Information Science, 13 (4): 533–555
Bian F L, Sha Z Y, Chen J P (2001) Generation of minimum rules from rough rule sets based on object-spatial information table.Geomatics and Information Science of Wuhan University, 26(5): 399–403
Rymon R (1992) Search through systematic set enumeration. The 3rd International Conference on Principle of Knowledge Representation and Reasoning. Massachusetts, USA.