Title :
A linguistically sound approach to content analysis of natural language text
Author :
Hafner, Carole D.
Author_Institution :
Coll. of Comput. Sci., Northeastern Univ., Boston, MA, USA
Abstract :
An approach to text content analysis aimed at utilizing linguistic structure effectively within a robust, pattern-directed framework is described. The method, which is under development as part of the ROBIN (robust interpreter) text analysis system, uses a bottom-up chart parser to collect information about the linguistic structure of the input. The chart is then subjected to rule-based pattern matching to recognize the concepts of interest. It is shown how the structure-matching approach is able to use syntactic information to distinguish correct from incorrect examples of a concept, even when the chart provides only a partial analysis of the input
Keywords :
character recognition; linguistics; natural languages; word processing; ROBIN; bottom-up chart parser; content analysis; linguistic structure; linguistically sound approach; natural language text; partial analysis; pattern-directed framework; robust interpreter; rule-based pattern matching; structure-matching approach; syntactic information; text analysis system; text content analysis; Computer science; Educational institutions; Information analysis; Natural languages; Pattern analysis; Pattern matching; Pattern recognition; Robustness; Text analysis; Text recognition;
Conference_Titel :
AI Systems in Government Conference, 1990. Proceedings., Fifth Annual
Conference_Location :
Washington, DC
Print_ISBN :
0-8186-2044-7
DOI :
10.1109/AISIG.1990.63814
