Parsing Query Interfaces of Deep Web from Specialization to Generalization

E-commerce Web sites provide the economic information by query interfaces. Query interfaces are considered as deep Web services that indirectly reflect the real schema of the hidden databases. The valid accessing method of deep Web is through query interfaces. However, the query interfaces are designed and coded in different development models and using different programming languages and the semantic relation is very complicated. To improve the accuracy of extracting attributes, an approach is presented to extract the attributes heuristically. During the extraction, attributes are enriched utilizing ontology by which we get deep semantic understanding of query interfaces. Experiments over real-word domains show the utility and ability of the algorithm for grasping the information of query interfaces.