Name
Title | ||
|---|---|---|
Learning geoscience categories in Situ: implications for geographic knowledge representation |
Abstract | ||
|---|---|---|
This paper explores the development of categories shared in the field logging of a region by a team of geologists. Visualization, neural networks and spatial statistical tools are employed to gain insight into the complex space of attributes observed, and into the categories developed. Background material and a discussion of results examines the findings in the light of research into category development, and specifically how categories are thought to be formed and modified as part of the (geo)scientific process and the situations encountered. Results show that (1) category discrepancy exists between individuals; (2) category development or revision is evident among individuals; and (3) that some categories do not seem to be totally defined by observed data alone. The results imply that contextual factors should also be considered when adopting ontological approaches to information representation. |
Year | DOI | Venue |
|---|---|---|
2001 | 10.1145/512161.512190 | ACM-GIS |
Keywords | Field | DocType |
ontological approach,observed data,neural network,geographic knowledge representation,contextual factor,background material,complex space,geoscience category,information representation,category discrepancy,category development,scientific process,self organizing maps,classification,spatial statistics,situated learning,knowledge representation | Ontology,Data mining,Knowledge representation and reasoning,Visualization,Computer science,Self-organizing map,Artificial intelligence,Situated learning,Artificial neural network,Machine learning,Information representation,Scientific method | Conference |
ISBN | Citations | PageRank |
1-58113-443-6 | 8 | 0.83 |
References | Authors | |
5 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Boyan Brodaric | 1 | 129 | 14.10 |
| Mark Gahegan | 2 | 571 | 55.38 |