Completing Queries: Rewriting of Incomplete Web Queries under Schema Contraints

Web queries have been and will remain an essential tool for accessing, processing, and, ultimately, reasoning with data on the Web. With the vast data size on the Web and Semantic Web, reducing costs of data transfer and query evaluation for Web queries is crucial. To reduce costs, it is necessary to narrow the data candidates to query, simplify complex queries and reduce intermediate results. This article describes a static approach to optimization of web queries. We introduce a set of rules which achieves the desired optimization by schema and type based query rewriting. The approach consists in using schema information for removing incompleteness (as expressed by ‘descendant’ constructs and disjunctions) from queries. The approach is presented on the query language Xcerpt, though applicable to other query languages like XQuery. The approach is an application of rules in many aspects—query rules are optimized using rewriting rules based on schema or type information specified in grammar rules.