Understanding Location Descriptions in the LEI System
40933Authors
nan
Abstract
Biological specimens have historically been la- beled with English descriptions of the location of collection. To perform spatial, statistical, or historic studies, these descriptions must be converted into geodetic coordinates. A study of the sublanguage used in the descriptions shows much less frequent than typical usage of observer-relative relations such as "left of," but shows problems with name ambiguity, finding the referents of generic terms like "the stream," ordinal numbering of river forks and valley branches, object-oriented prepositions ("be- hind"), fuzzy boundaries (how close is "at," how far is still "north of"), etc. The LEI system implements a semi-automated understanding of such location descriptions. Components of LEI include a language analyzer, a geograph- ical reasoner, an object-oriented geographic knowledge base derived from US Geological Survey digital maps with user input, and a graphical user interface. LEI parses preposi- tional phrases into spatial relations, converts these into areas, then computes polygon over- lays to find the intersection, and returns the minimum bounding rectangle. The user is con- suited on unknown words/phrases and ambigu- ous descriptions. 1
Keywords
Full Content
Understanding Location Descriptions in the LEI System David N. Chin Matthew McGranaghan Tung-Tse Chen Dept. of Info. & Computer Sciences Dept. of Geography Dept. of Info. & Computer Sciences University of Hawaii University of Hawaii University of Hawaii 2565 The Mall 2424 Maile Way, Porteus Hall 445 2565 The Mall Honolulu, HI 96822 Honolulu, HI 96822 Honolulu, HI 96822 Chin@Hawaii .Edu matt@uhunix, uhcc. hawaii .edu Abstract Biological specimens have historically been la- beled with English descriptions of the location of collection. To perform spatial, statistical, or historic studies, these descriptions must be converted into geodetic coordinates. A study of the sublanguage used in the descriptions shows much less frequent than typical usage of observer-relative relations such as "left of," but shows problems with name ambiguity, finding the referents of generic terms like "the stream," ordinal numbering of river forks and valley branches, object-oriented prepositions ("be- hind"), fuzzy boundaries (how close is "at," how far is still "north of"), etc. The LEI system implements a semi-automated understanding of such location descriptions. Components of LEI include a language analyzer, a geograph- ical reasoner, an object-oriented geographic knowledge base derived from US Geological Survey digital maps with user input, and a graphical user interface. LEI parses preposi- tional phrases into spatial relations, converts these into areas, then computes polygon over- lays to find the intersection, and returns the minimum bounding rectangle. The user is con- suited on unknown words/phrases and ambigu- ous descriptions. 1 Introduction Many biological specimens collected in the past 1 are la- beled with only an English description of their location of collection. 2 To perform any statistical or spatial anal- ysis of this historical data, these descriptions must be converted into geodetic coordinates (latitude-longitude or UTM), a time-consuming process that requires eye- straining poring over maps to search for each location. 1Current collectors can use hand-held satellite-based geo- positioning systems to record collection coordinates. 2There are an estimated several hundred millions of such labeled specimens. To automate this process requires understanding the nat- ural language descriptions, reasoning about the spatial re- lations described by the natural language, and mapping these into a geographical object base to derive the collec- tion coordinates. 2 Previous Research Talmy [1983], Herskovits [1986], and Andr6 etal. [1986] among others have documented the many problems in in- terpreting and using spatial prepositions. For example, in and on have similar but different meanings: "in the car" means within the car, while "on the car," means on top of the car. However, "on the bus/plane," means within the bus or plane. Also, each preposition typically has several different meanings or usages. For example, one says "at home," but "at the bank," and the meaning of "the plane is at Honolulu airport," is within the area of Honolulu air- port, but the meaning of "the dog is at the telephone pole" is not within the telephone pole, but near it. These context dependent usages make interpretation and application of spatial prepositions problematic. Kuipers [1985], Davis [1986], Peuquet and …
Content truncated to 500 words
Labeled Text
Biological specimens have historically been la- beled with English descriptions of the location of collection. To perform spatial, statistical, or historic studies, these descriptions must be converted into geodetic coordinates. A study of the sublanguage used in the descriptions shows much less frequent than typical usage of observer-relative relations such as "left of," but shows problems with name ambiguity, finding the referents of generic terms like "the stream," ordinal numbering of river forks and valley branches, object-oriented prepositions ("be- hind"), fuzzy boundaries (how close is "at," how far is still "north of"), etc. The LEI system implements a semi-automated understanding of such location descriptions. Components of LEI include a language analyzer, a geograph- ical reasoner, an object-oriented geographic knowledge base derived from US Geological Survey digital maps with user input, and a graphical user interface. LEI parses preposi- tional phrases into spatial relations, converts these into areas, then computes polygon over- lays to find the intersection, and returns the minimum bounding rectangle. The user is con- suited on unknown words/phrases and ambigu- ous descriptions. 1
Publication Details
- Paper ID:
- 40933
- Year:
- 1994
- Publication:
- TIPSTER
- Publication Type:
- conf
- DOI:
- nan