Abstract

Abstract. Peer-to-Peer (P2P) data management systems combine traditional schema-based integration techniques with the P2P infrastructure. In this paper, we propose a P2P data management framework named PEPSINT that semantically integrates heterogeneous XML and RDF data sources, using a hybrid architecture and a global-as-view approach. Our focus is on the query processing techniques over heterogeneous data. Queries in PEPSINT are expressed in XQuery and in RDQL. We consider two types of queries, depending on whether the query is first posed on the super peer or on one of the peers.

In this paper, we further develop a proposed layered approach for the Semantic Web. Our objective is to build a specific solution to the problem of providing data interoperability among different databases, so as to allow for schematic data integration. In particular, we solve the problem of translating queries on a database schema into queries on another database schema, using their relationship with an ontology. We use RDF Schema to model the databases and the ontology. A common vocabulary expresses the mappings between each database schema and the ontology.

Abstract. While providing a uniform syntax and a semistructured data model, XML does not express semantics but only structure such as nesting information. In this paper, we consider the problem of data integration and interoperation of heterogeneous XML sources and use an ontology-based framework to address this problem at a semantic level. Ontologies are extensively used for domain knowledge representation, by virtue of their conceptualization of the domain, which carries explicit semantics. In our approach, the global ontology is expressed in RDF Schema (RDFS) and constructed using the global-as-view approach by merging individual local ontologies, which represent XML source schemas. We provide a formal model for the mappings between XML schemas and local RDFS ontologies and those between local ontologies and the global RDFS ontology. We consider two cases of query processing, specifically for data integration and for data interoperation. In the first case, the user poses an RDF query on the global ontology, which is answered using all the mapped XML sources. In the second case, a query is posed on a single source and then is mapped to the XML sources that are connected to that source. For each case, we discuss the problem of query containment and present an equivalent query rewriting algorithm for queries expressed in two languages: conjunctive RDQL and conjunctive XQuery.

We propose a layered framework for the integration of syntactically, schematically, and semantically heterogeneous networked data sources. Their heterogeneity stems from different models (e.g., relational, XML, or RDF), different schemas within the same model, and different terms associated with the same meaning. We use a semantic based approach that uses a global ontology to mediate among the schemas of the data sources. In our framework, a query is expressed in terms of one of the data sources or of the global ontology and is then translated into subqueries on the other data sources using mappings based on a common vocabulary. Metadata representation, global conceptualization, declarative mediation, mapping support, and query processing are addressed in detail in our discussion of a case study.