Knowledge discovery for dealing with the empty query problem in relational database systems

An algorithm, CONEQ, is presented aimed at filtering out certain types of empty queries made to a relational database system. This algorithm employs the existing set of semantic integrity constraint rules for the database together with the current database state in a typical generate-and-test cycle aimed at creating a stronger set of semantic integrity rules that better reflect the particular database state. The algorithm is sound but not complete and a discussion of its implementation in Prolog is given.