Digital Science Popularization Resources Classification System and Metadata Exchange Study
|School||Central China Normal University|
|Keywords||science popularization digital science popularization resources resourceclassification metadata metadata conversion|
The digital information era, network science popularization is not only the important characteristic of modern science, but also an important way to promote popular science resource sharing. Network of popular science, however, there are also many problems, example:duplication of resources construction significantly (especially at the local science website), resource normative, low utilization; the original capacity is weak, a large total number, a small number of high-quality specialty resources; most of the site resource evaluation as a standard size or amount of resources, lack of content of the resource evaluation; lack of the concept of resource sharing with limitation on traditional portal and benefits etc.. To solve the above problem, this study expanded from the classification and standardization of construction of digital science popularization resources.First of all, on the basis of the advantages and disadvantages of the main classification methods at home and abroad, combining the existing classification of popular science resources, digital resources, learning resources, natural science and social science resources, with the characteristics of digital science popularization resources, mainly to the theme classification method, established a relatively complete and good scalability classification system of digital science popularization resources. The definition and characteristics of digital science resources were analyzed.Secondly, analyzed the advantages and disadvantages of the current domestic and foreign main metadata standards, combining with the characteristics of digital science popularization resources, selected DC as the primary reference standard; In accordance with the classification system of digital science popularization resources, developed the digital science popularization resources metadata standards, including digital science popularization resource metadata core elements set, audio and video general metadata, images general metadata, the literature/text general metadata, experience general metadata, and exhibition, reports/seminars, match general metadata. Digital science resources images metadata as an example described the digital science popularization resources metadata standard with XML/RDF.Metadata conversion part, adopted artificial matching+program transformation+manual patching way to achieve, namely:artificial definition the rules of metadata matching, procedures for automatic transformation, artificial reading and repairing, so as to maximize the guarantee the matching accuracy and conversion efficiency, reduce conversion time and data loss rate. Artificial matching part, used the "directory" element coding, an innovative way to simplify the issues of metadata span structure, multilevel conversion; program conversion section, selected DC as the intermediary model, and used DOM to implement the conversion of metadata xml file, which based on the label (attribute value and the range)(with digital science popularization resource images metadata for instance); Artificial repair part, mainly for a few fuzzy matching elements are read and modified to accurate. Selected20samples, to validate the conversion method with precision P, full-rate R and F values.Finally, this paper research summary and improved prospects.