System Entity Structure For XML Meta Data Modeling; Application to the US Climate Normals

摘要

Data engineering becomes increasingly important with the increasing popularity of service oriented architecture (SOA) and web services. Such data engineering requires a design methodology within an ontology framework. In this paper, the System Entity Structure (SES) serves as an abstract ontology framework for world state descriptions particularly involving dynamics in space and time. It is implemented within a software tool, the SESBuilder, employing the extended markup language XML. Pruned Entity Structures (PES) represent the logically possible set of world state descriptions consistent with the SES. At the implementation level, an SES is represented by a schema or DTD whose instance documents represent possible prunings. The SESBuilder supports convenient specification of SESs, pruning to create PESs, and transformation to XML representations, all through natural language and graphical interfaces. Thus the software provides a capable data engineering work space. In this paper, we illustrate its application to data modeling of US Climate Normals for the purpose of automated generation of weather simulation models.