Research On Automating Heterogeneous XML Transformation

With the rapid development of computer and network technology, XML has the characteristics of self-description、extension which makes XML itself is very suitable for data description in several application domains. However, different application system is increasing quickly, which results in a large number of heterogeneous XML data. In order to shield the structure and environment configuration of the different data source, we often use XML as intermediate standard language. Through translate the variety of heterogeneous data sources into a unified XML standard intermediate language, which make it easy to share the data between heterogeneous systems. Hence, automate transformation of XML data is regarded as an important activity in heterogeneous systems with any data model, where data structures of one format must be transformed into a form accepted by another one. This is typically attained in real world by writing specific transformation languages which requires professional programming skills. Especially, when data is in large scale and frequently changing, the manual transformation is an extremely time-consuming and error-prone process. Therefore, automate as much as possible all steps of XML document transformation process is of great significance, This paper compare and analyze the main technology of XML transformation, then proposed a method automatically transformation XML document based on schema matching. The work of this paper mainly includes the following aspects:(1) This paper analysis the development situation about XML transformation technology and describe the approaches of schema matching in detail. And the overall design framework of XML transformations based on schema matching is proposed.(2) The visualization techniques is introduced to XML Schema modeling process, this paper model an XML schema as a directed label graph with constraint sets. Through the visual user interface makes it easier for users to understand the structure of XML Schema.(3) This paper proposes an XML schema matching algorithm. First a name is tokenized into a list of words, and we combine semantic similarity and edit distance to compute name similarity. Then we compute structure similarity by consideration of data types and the context of element. Finally, the user adjusts the proposed match result and obtains the precise matches between the two schemas. The algorithm takes into account the full syntax and semantics of information, structure information, greatly improving the matching quality. Experimental of several real applications show encouraging results, yield high precision and recall measures. (4) We define a set of transformation operations used to solve the schema conflict between semantics and structure. In order to save to mapping results in standard format, we describe mapping rules definition, and save the mapping relationship in the form of XML file. The role of the mapping result is to semantically relate facts from the source and target schemas by encapsulating all necessary information to transform instances of one source schema to instances of one target schema.(5) Based on XSLT engine, the mapping results can be generates XSLT script that supports the transformation of XML documents.The research on automatically transformation of XML between heterogeneous applications is quite a new research field. This dissertation has made some tentative approaches to the XML automatic transformation in the field of PLC graphic programming domain, which establishes good foundation for the data exchanging cross platform. Some theoretical outcome has been achieved which could serve as the theoretical foundation of the similar software research and development in many other domains.