Research And Application Of Heterogeneous Data Integration Based On XML And Web Services

Along with the development and popularization of network and distributed application, more and more heterogeneous data sources had been produced; the need to solve the integration problem of these heterogeneous data sources was becoming more important. The purpose of data integration system is to provide users with seamless and flexible access to heterogeneous data source through a unified query platform which can hide the difference of underlying data sources.The paper focuses on relevant theories and technologies of heterogeneous data integration and presents an extended architecture called HDQAS (Heterogeneous Data Query and Analyzing System), which is based on analysis of the current data integration approaches and Mediator/Wrapper method as well as data warehouse. HDQAS adopts XML technique to solve the problems of syntax heterogeneity. This structure uses the advantage of ontology to describe precise semantic relationships, and uses OWL (Ontology Web Language) to represent global ontology and local ontology. Local ontology is the ontology description of specific data source and global ontology is an overall view of local ontology. It solves the problems of semantic heterogeneity through mapping between global ontology and local ontology. HDQAS masks the platform discrepancy of data sources based on Web services technology, implements data integration in a larger range through extracting, encapsulating and invoking of various data sources especially Web pages. In order to provide the decision support to the customers, HDQAS carries on further analysis of the result of data integration with warehouse and OLAP technology.At the end of the paper, a heterogeneous data integration prototype system under the HDQAS frame is realized. The client of the prototype provides a unified query interface through importing of global ontology, and can perform data query and on-line analytical processing. The server carries out functions such as the registration and selection of data sources, wrapper maintenance and user management.The main achievements of this paper are as follows:1. The heterogeneous data integration architecture is designed. The functions of components are designed in detail, and the key modules of the prototype system are realized.2. The method of wrapping data source to construct Web Services features the...