Querying Relational Databases without Explicit Joins

Despite its benefits and wide-spread acceptance, SQL [5] is not a perfect query language. Although graphical tools for query construction mask some of the complexity, complex database schema challenge even experienced database users during query formulation because a user is responsible for mapping the semantics of their query to the structure of the database. In this work, we propose a semantic query language for graphically querying relational database systems that allows a user to query the database by semantics instead of structure. Database semantics are described using a global dictionary and semantic specifications that are combined to form an integrated, context view. Users query the semantic view by concept name, and the query processor translates semantic queries to SQL. This translation involves automatically determining attribute and relation mappings and join conditions.