Share Email Print

Proceedings Paper

Quicky location determination based on geographic keywords of natural language
Author(s): Danhuai Guo; Weihong Cui
Format Member Price Non-Member Price
PDF $14.40 $18.00
cover GOOD NEWS! Your organization subscribes to the SPIE Digital Library. You may be able to download this paper for free. Check Access

Paper Abstract

In location determination based on natural language, it is common to find the location by describing relationship between the undetermined position and one or several determined position. That indicates that the uncertainty of location determination processing is derived from the one of natural language procedure, the one of spatial position description and the one of spatial relationship description. Most of current researches and regular GIS software take certainty as prerequisite and try to avoid uncertainty and its influence. The research reported in this paper is an attempt to create a new combing method of Artificial Intelligence (AI), Fuzzy set theory and spatial information science named Quickly Location Determination based on Geographic Keywords (QLDGK) to rise to the challenge of location searching technique based on natural language. QLDGK have two technical gists. The first one is geographic-keywords-library and special natural-language-separation-model-library that increases the language processing efficiency. The second one is fuzzy theory based definition of spatial relationship, spatial metric and spatial orientation that extends the searching scope and defines variant confidences on variant searching outcome. QLDGK takes consideration on both higher query efficiency and the lower omission rate. The above method has been proved workable and efficient by QLDGK prototype system which was tested by about 12000 emergency call reports from K-city, Southwest of China, and achieved the test result with 78% accuracy in highest confidence and 8% omitting ration.

Paper Details

Date Published: 4 September 2007
PDF: 13 pages
Proc. SPIE 6754, Geoinformatics 2007: Geospatial Information Technology and Applications, 67540U (4 September 2007); doi: 10.1117/12.764634
Show Author Affiliations
Danhuai Guo, Institute of Remote Sensing Applications (China)
Weihong Cui, Institute of Remote Sensing Applications (China)

Published in SPIE Proceedings Vol. 6754:
Geoinformatics 2007: Geospatial Information Technology and Applications

© SPIE. Terms of Use
Back to Top