A system for query comprehension

Abstract

A major difficulty of query writing is that users do not fully understand queries. A system providing feedback is proposed so as to enable users to better understand their queries. It is a multi-functional rule-based system. In addition to providing an English translation for the query, it pin-points syntax errors, and suggests possible solutions. It also finds likely semantic errors, and provides possible solutions. The system was implemented for a query language based on the entity relationship model. It is expected to lead to better user performance. Empirical testing showed significant improvement in user performance with the feedback system. The system can be used for computer-aided training and in actual operations.