Encoding heuristic knowledge for GIS classification

Author

Kirkby, S.D. ; Badcock, J. ; Eklund, P.W.

Author_Institution

Nat. Key Centre for Social Appl. of GIS, Adelaide Univ., SA, Australia

Volume

2

fYear

1997

fDate

28-31 Oct 1997

Firstpage

1143

Abstract

This paper discusses a sequence of experiments comparing classification results produced by geographic information systems (GIS) to those produced by integrated GIS/expert systems (ES). Analysis is conducted via the use of three machine learning techniques: inductive learning, instance based learning and neural networks. The paper argues that inferior knowledge engineering techniques result in poor heuristic knowledge encoding in both traditional GIS and more recent GIS/ES classifications

Keywords

classification; expert systems; geographic information systems; heuristic programming; knowledge engineering; learning (artificial intelligence); neural nets; GIS; classification; experiments; expert systems; geographic information systems; heuristic knowledge encoding; inductive learning; instance based learning; knowledge engineering; machine learning; neural networks; Australia; Computational Intelligence Society; Encoding; Engines; Expert systems; Geographic Information Systems; Information analysis; Knowledge engineering; Machine learning; Neural networks; Testing;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Processing Systems, 1997. ICIPS '97. 1997 IEEE International Conference on

Conference_Location

Beijing

Print_ISBN

0-7803-4253-4

Type

conf

DOI

10.1109/ICIPS.1997.669164

Filename

669164