Font Size: a A A

Mapping Integrity Constraint Ontology To The Relational Database

Posted on:2012-09-23Degree:MasterType:Thesis
Country:ChinaCandidate:X J CuiFull Text:PDF
GTID:2178330332499580Subject:Computer software and theory
Abstract/Summary:PDF Full Text Request
Semantic Web is the extension of the current Web, in which information is expressed in intelligent forms, better enabling computers and people to work in cooperation. As the basis of the Semantic Web, ontology provides a source of shared and precisely defined terms that can be used in description of Web source and OWL is a well-known language for ontology modeling in the Semantic Web. RDF is the W3C recommended standard data model that describes Web resource in triple forms, using SPARQL, the query language for RDF that has been recommended by W3C, to query the Semantic Web Data.Semantic Web has achieved success in many applications, such as information retrieval and data integration, etc. However, it is still a long process for real data-centric applications. On the other hand, since the vast majority of data on the World Wide Web are still stored in the database, how to use the existing mechanisms to store Semantic Web data in the relational database is an important issue. Motivated by the incompleteness problem encountered in mapping ontology to the relational database, in this paper, we implement the mapping from the point of integrity constraint validation, storage mechanism and query translation by comparing the axioms of ontology with the data schemas of the relational database.In this paper, the IC-axioms have been defined to represent the integrity constraint in ontology. Generally, the integrity constraint validation need to be implemented in closed world semantics. However, the axioms of ontology are used for checking the consistency of ontology and inferring implicit facts, and it is more applicable in open world reasoning, which makes the integrity constraint of ontology encountered difficulties in validation. Based on this, the axioms distinguish the IC-axioms that used for checking the satisfaction of integrity constraints under closed world assumption from the standard axioms that imply new information under open world assumption. In further, DCQnot is used for the integrity constraint validation. In this validation, the IC-axioms are converted into the DCQnot and further converted to the corresponding SPARQL queries. In this way, the integrity constraint validation is transferred into the query answering.Currently, the storage mechanisms of Semantic Web data focus on the simple triple table storage and property partition table storage, and improve the query efficiency mainly through increasing index and compression, but not directly face to the schema integration problem. According to the IC-axioms, we decompose the schemas and create several different types of tables, then store the ontology data to the corresponding tables, which optimize the existing storage approaches. This storage is more suitable for people to understand and deal with the data. Comparing with the existing storage, the experimental result shows that our storage is superior to the traditional approaches in the majority case, and with increasing the dataset capacity, the discrepancy is in evidence.The current mainstream query translations are mainly based on the traditional storages, such as triple tables or property partition tables, and it is not applicable to our storage and further has poor readability. Based on this, we classify the SPARQL query statements according to the tables generated by our storage and further proposed the corresponding translation algorithms of SPARQL to SQL. The generated SQL statements are rather simple, easy to understand. In further they have less table join, which making the query more efficient.Since the Semantic Web has been proposed, it has been a discussion and research focus to the experts and scholars. Studying how to store and query the Semantic Web data in relational database under the premise of maintaining its semantic is a current research focus.
Keywords/Search Tags:Semantic Web, Ontology, Integrity Constraint, Relational Database, Description Logic, SPARQL
PDF Full Text Request
Related items