Formal Verification of Semistructured Data Models in PVS

Abstract

The rapid growth of the World Wide Web has resulted in a dramatic increase in semistructured data usage, creating a growing need for effective and efficient utilization of semistructured data. In order to verify the correctness of semistructured data design, precise descriptions of the schemas and transformations on the schemas must be established. One effective way to achieve this goal is through formal modeling and automated verification. This paper presents the first step towards this goal. In our approach, we have formally specified the semantics of the ORA-SS (Object-Relationship- Attribute data model for Semistructured data) data modeling language in PVS (Prototype Verification System) and provided automated verification support for both ORA-SS schemas and XML (Extensible Markup Language) data instances using the PVS theorem prover. This approach provides a solid basis for verifying algorithms that transform schemas for semistructured data.