A SUBLANGUAGE APPROACH TO NATURAL-LANGUAGE PROCESSING FOR AN EXPERT-SYSTEM

A sublanguage grammar approach with strong reliance on semantic word c lasses was used to develop an NLP component for processing the free-te xt comments on life insurance applications for evaluation by an underw riting expert system. The NLP component is implemented using a logic-p rogramming formalism. Lexicon entries contain semantic word class tags such as BODYPART, SYMPTOM, or MEDICATION. Adjacency Rules and an Ambi guity Filter are used to interpret the input data using these semantic word classes. Across two experiments, the system produced appropriate readings for 96.8% of a sample of 2069 utterances, and the number of parses produced per utterance was 1.079 for the same sample. Misspelli ngs caused the system its only serious problem. The sublanguage approa ch to processing text was shown to be very promising for expert system s and suggests itself as a useful paradigm for a range of other text-b ased systems which must deal with naturally occurring and frequently u ngrammatical texts.