Abstract
Imprecise composite location references formed using ad hoc spatial expressions in English text makes the geocoding task challenging for both inference and evaluation. Typically such spatial expressions fill in unestablished areas with new toponyms for finer spatial referents. For example, the spatial extent of the ad hoc spatial expression "north of" or "50 minutes away from" in relation to the toponym "Dayton, OH" refers to an ambiguous, imprecise area, requiring translation from this qualitative representation to a quantitative one with precise semantics using systems such as WGS84. Here we highlight the challenges of geocoding such referents and propose a formal representation that employs background knowledge, semantic approximations and rules, and fuzzy linguistic variables. We also discuss an appropriate evaluation technique for the task that is based on human contextualized and subjective judgment.
Highlights
A geographic coordinate system, such as the World Geodetic System 1984 (WGS84), quantitatively describes the exact area of named locations
Human qualitative spatial referents appear as both established toponyms (e.g., "Dayton, Ohio") or ad hoc spatial expressions (e.g., "north of Dayton, Ohio") linked to established/atomic toponyms with relations
We identify and operate on polygons to represent qualitative spatial referents
Summary
A geographic coordinate system, such as the World Geodetic System 1984 (WGS84), quantitatively describes the exact area of named locations. While a lat/long point on a map, as in WGS84 has precise and distinctive semantics, adding linguistic expressions such as "north of" or "not far from" to that point makes it difficult for spatial systems to infer the exact spatial extent of the referent because the area is not precisely defined. This requires translation from a qualitative representation to an (imprecise) quantitative one that is adequate for the intended application and context. We suggest evaluating proposed representations using a measurement scale with a higher resolution relative to a binary decision (the Likert scale), to collect human judgment and better capture the intuition behind the use of ad hoc spatial expressions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
