Formalizing natural-language spatial relations descriptions with fuzzy decision tree algorithm

Jun Xu; Changqing Yao

doi:10.1117/12.712745

28 October 2006 Formalizing natural-language spatial relations descriptions with fuzzy decision tree algorithm

Jun Xu, Changqing Yao

Proceedings Volume 6420, Geoinformatics 2006: Geospatial Information Science; 64201E (2006) https://doi.org/10.1117/12.712745
Event: Geoinformatics 2006: GNSS and Integrated Geospatial Applications, 2006, Wuhan, China

Abstract

People usually use qualitative terms to express spatial relations, while current geographic information systems (GIS) all use quantitative approaches to store spatial information. The abilities of current GIS to represent and query spatial information about geographic space are limited. In order to incorporate the concepts and methods people use to infer information about geographic space into GIS, research on the formal model of common sense geography becomes increasingly important. Previous research on the formalizations of natural-language descriptions of spatial relations are all based on crisp classification algorithms. But the human languages about spatial relations are ambiguous. There is no clear boundary between "yes" or "no" if a spatial relation predicate can express the spatial relations between objects. So the results of crisp classification algorithms can not formalize natural-language terms well. This paper uses a fuzzy decision tree method to formalize the spatial relations between two linear objects. Topologic and metric indices are used as variables, and the results of a human-subject test are used as training data. The formalization result of the fuzzy decision tree is compared with the result of a crisp decision tree.

Citation Download Citation

Jun Xu and Changqing Yao "Formalizing natural-language spatial relations descriptions with fuzzy decision tree algorithm", Proc. SPIE 6420, Geoinformatics 2006: Geospatial Information Science, 64201E (28 October 2006); https://doi.org/10.1117/12.712745

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available