A Common Architecture to Encourage Reuse of Natural Languagmext Processing Tools (long version)
T. Richard MacMillan, Elaine P. Lusher, Martha L. Farinacci, Sharon J. Laskowski, Leonard J. Seligman, Adrienne J. Kleiboemer · 1993
Text analysis system that handle large amounts offreeform narrative are beginning to use natural language processing algorithms, employing linguistic-based techniques to deduce meaning. Existing natural languugdtext processing (NLJTP) systems often serve similar purposes and use similar components, but component reuse is rare. We have designed a domain-based architecture to promote reusability of NUTP software and data components. It is constructed from an ontology of the NLJTP domain, which was derived from a survey of NVTP applications and users, and is represented by an objectoriented model. The architecture also includes interoperability standards, inte$ace protocols, and tools for developing reusable components intended to promote reuse of software and data across a wide range of NLJTP applications and technologies, from keyword search to semantic analysis.