Developing a Domain Specific Geospatial Feature Type Ontology for a GeoReferenced Information Portal (GRIP)
Douglas M. Flewelling, Ph.D.
Mark P. Kumler, Ph.D.
In this project, geospatial ontology is used in the process of information retrieval, both for answering and deducing the user queries, and for semantically organizing the query results. The purpose of this geospatial ontology is to extract and aggregate the knowledge about desert tortoises and to provide a standard vocabulary to the users, which will facilitate the process of information retrieval and extraction from the desert tortoise database repository. The ontology is based on the most commonly asked questions and most frequently used terms about desert tortoises. The Algernon inference engine is used in this project to answer the various queries due to its ability to store the results of queries in the form of tables. This project provides the starting point, an initial guide to help the desert tortoise research community further develop the knowledge related to desert tortoise recovery plans.
The main goal of this project was to bring together many different types of information in a way that allows all the data to be used together so that a user can query the data and retrieve the information spatially and textually. This was accomplished by organizing the information semantically and hierarchically in terms of various relations among different classes as well as attributes of those categories. The use of ontology facilitates the new information discovery, which otherwise is not explicitly stated in the repository. The geo-ontology developed in this project will play a key role in distinguishing the presence of a place name in a query, formulating the system queries, extracting metadata, and organizing search results. Additionally, the ontology for a particular domain for a particular user community provides them with a common vocabulary which, in turn, helps to reduce ambiguity and misunderstanding.
Bansal, N. (2007). Developing a Domain Specific Geospatial Feature Type Ontology for a GeoReferenced Information Portal (GRIP) (Master's thesis, University of Redlands). Retrieved from http://inspire.redlands.edu/gis_gradproj/101
Full text is available at the University of Redlands