Based On The Body Of Heterogeneous Data Source Integration System Model And Its Query Processing

Web ofers an abundant and valuable information repository. So it is an important and meaning subject to research how discovery and obtain valuable information in Web repository. The data of Web has characteristics such as semistructured, distributed, heterogeneous etc, so it is difficult for traditional database technique to storage and query web source. In this paper, we focus on the study of integration and querying of heterogeneous data sources.In this paper, we propose an approach for integration of heterogeneous data sources and query processing across these data sources, on the basis of analysing the characteristics of the web and the technique of information integration for web. In brief,we make the following contributions in this paper:1. We provide an ontology-based framework OBISUA (Ontology-Based Integration System Universal Architecture) for integration and querying Web-accessible heterogeneous data sources. This framework resolve the problem of semantic heterogeneous between different data sources and provide a uniform query interface for the user.2. Two efficient algorithms are proposed for rewriting user queries. One rewrite a query on the global ontology to queries on the local ontology, the other rewrite a query on the local ontology to queries over the data sources. We use RDFS to model the ontologies, use RDQL to query global ontology, XQuery to query data sources.3. In order to reduce responding time for query, we introduce cache to save the results of prior queries. Use LRU algorithm delete the results that use less. In order to judge whether the result of the query included in the cache, we propose the concepts of "cache entry" and "query relation vector". Cache reduce responding time for query and reduce the pressure of internet and query power of data sources.