Spatial data quality capture through inductive learning

Authors: Duckham M.¹; Drummond J.²; Forrest D.²

Source: Spatial Cognition and Computation, Volume 2, Number 4, 2000 , pp. 261-282(22)

Publisher: Springer

Buy & download fulltext article:

Price: $47.00 plus tax (Refund Policy)

Abstract:

The relatively weak uptake of spatial error handling capabilities by commercial GIS companies and users can in part be attributed to the relatively low availability and high costs of spatial data quality information. Based on the well established artificial intelligence technique of induction, this paper charts the development of an automated quality capture tool. By learning from example, the tool makes very efficient use of scarce spatial data quality information, so helping to minimise the cost and maximise availability of data quality. The example application of the tool to a telecommunications legacy data capture project indicates the practicality and potential value of the approach.

Keywords: data quality; GIS; inductive learning; information content; object calculus

Language: English

Document Type: Regular paper

Affiliations: 1: Department of Computer Science, University of Keele, ST5 5BG, UK (Fax: +44 1782 713082; E-mail: m.duckham@computer.org) 2: Department of Geography and Topographic Science, University of Glasgow, G12 8QQ, UK

Publication date: 2000-01-01