Semantics modeling issues for processing natural language database queries

  • Authors:
  • D. G. Shin

  • Affiliations:
  • Computer Science and Engineering Dept., University of Connecticut, Storrs, CT

  • Venue:
  • CSC '90 Proceedings of the 1990 ACM annual conference on Cooperation
  • Year:
  • 1990

Quantified Score

Hi-index 0.00

Visualization

Abstract

This work explores data semantics issues necessary for designing an NL interface capable of dealing with incomplete NL queries. First, the notion of incomplete NL queries is clarified to conclude the necessity for enriching the data semantics component of the system. An assumption made by the existing NL interfaces is expounded to argue that the approaches based on this assumption are unable to explore meta-knowledge about the stored data in the query conversion process. A new, unconventional approach is proposed whose tenets underlie the separation between the conceptual interpretation of NL queries and the semantics descriptions of the database constituents. This approach allows explicit descriptions of the conceptual knowledge, referred to herein as epistemological specifications, on the DB constituents.