Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Ontology-assisted Information Extraction in Agricultural Publications

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  Paper number  033062,  2003 ASAE Annual Meeting . (doi: 10.13031/2013.14049) @2003
Authors:   Soonho Kim, Howard W. Beck
Keywords:   Ontology-assisted information extraction system, crop-pest ontology, phrase patterns, Island chart parser, semantic structure, Ontology-assisted search

A crop-pest ontology was developed to organize and retrieve information in a crop-pest domain. We have developed an information extraction system that extracts information from text, as a first step toward constructing the ontology-assisted search. DDIS captions were used for developing the system. The crop-pest ontology provided a domain-specific knowledge source and phrase patterns that were used as a grammar to parse the captions in the system. Island chart parser parsed the captions based on the grammar. Parse trees generated as outputs of parsing were then converted to semantic structures, which can be used for retrieving information in an ontologyassisted search. Phrase patterns in the system were built manually. For the next phase of the project, we will examine ways to automatically acquire new patterns from existing patterns and input strings.

(Download PDF)    (Export to EndNotes)