A Paradigm For Natural Language Explanation Of Database Queries: A Semantic Data Model Approach

Although the increasing pervasiveness of end-user computing has made concerns with database interfaces increasingly more important, paradoxically, there does not seem to be sufficient research into database interfaces for end-user computing. A much needed but neglected enrichment of database systems is an interface that provides the user with automatic feedback in the form of an explanation of how the database management system interprets user-specified queries. The study presented in this paper proposes an approach that exploits the rich semantics of graphical semantic data models to construct restricted natural language explanations of database queries that are specified in a very high-level declarative form. These interpretations of the specified query represent the system’s “understanding†of the query, and are returned to the user for validation.