AN EXPERIMENT TOOL FOR THE MULTI-LAYERED GEOSCIENCE ONTOLOGY

One of the significant challenges towards the integration of geoscience datasets and query for the meaningful geological subjects, such as instances of the earth material types or deposits, from a virtual integrated geoscience database is the need to capture geoscience knowledge that geologists will be comfortable with. This paper reports on an experiment tool that supports compiling multi-layered ontology, or concept space as the community refer to as, and outputting geologic query concepts. The end users can browse the captured structural knowledge and query geological data in the integrated database via the geologic query concepts.

Basically, the tool supports to compile three kinds of geologic concept models. The first one is the hierarchical concept model. The hierarchical concept model defines what an independent geologic concept is and what the position it locates in the hierarchical concept model as well. The definition methods of the model are very similar to the classification schemes of geoscience terminology. The classification scheme depends on the geologic term to be classified. Thereafter, the definition methods form into the layers of the hierarchical model. The second one is the relational concept model. The relational concept model establishes relationships among the independent geologic concepts with a more structural style. And the last one is the query concept model. The query concept model builds up application oriented subjects that take the concept localization or specification into account.

The tool can output "geologic concept queries" that look like: "Select Terrigenous-clastic material.Mudrock in the area Lat-long area.areaA" or "Select Washover-fan deposit with the description of lithofaceA or lithofaceB."

Initially, the tool is designed for the Sedimentary materials: science language for their classification, description, and interpretation in digital geologic-map databases (NADMT, 2004). Therefore, almost all the interested terms in it could be arranged into one hierarchy under the top-level GeologicConcept model. In fact, it can capture geologic knowledge only if the geoscientists agree upon the representations predefined in the tool. We believe the tool will be upgraded when more geoscience knowledge to be captured.

Reference

North American Geologic-map Data Model Science Language Technical Team, 2004, Sedimentary materials: science language for their classification, description, and interpretation in digital geologic-map databases, Version 1.0 (12/18/2004): Draft report posted on the North American Data Model website , 595 p.