A Web-Enabled Approach for Generating Data Processors

Researchers in environmental sciences work with datasets that have a large variety of data structures and file formats. Consequently, for data and model interoperability, it is essential to have the right tools for converting from a given data structure to another, and from a specific file format to another. In this paper, we propose an original web-enabled approach for generating data processors capable of handling a multitude of data operations, including numerous data conversion and processing activities. The proposed approach emphasizes end-user, direct manipulation-based definition of data processors and automated code generation of their associated code, thus freeing the scientists from specialized and often tedious programming tasks. This, in turn, translates into an increased efficiency of the scientific research work. Details of the proposed approach and its supporting web-based software tool are presented in this paper, together with an application example in which an input dataset in the CSV format is converted into an output dataset in the XML format. The applicability of the proposed approach goes beyond the domain of environmental sciences, for which it was initially created, as it can be used in many other areas of research that emphasize data-intensive exploration and processing.