The instantiation of Omnipaper RDF prototype in the context of scientific publications

Abstract

The purpose of this paper is to present an instance of the system developed in the OmniPaper project, regarding the mechanisms of distributed information retrieval. These mechanisms were developed for newspapers’ articles and they were then instantiated in the context of the scientific publication. Another goal concerns the use of a central metadatabase developed to accomplish the syndication of contents, through the RSS approach. Design/methodology/approach One of the steps of the system’s development was the definition of the metadata layer that supports the research and the navigation functionalities as well as the contents’ syndication. Several tasks were performed for the definition of the metadata layer, namely: (1) analysis of several metadata standard vocabularies; (2) Selection of the metadata elements; (3) Definition of an application profile and the RSS template; (4) Development of a metadatabase, through the use of a native RDF database management system to store the RSS descriptions of the scientific publications; (5) Implementation of the search and navigation processes developed in the prototype through the use of the RDFS version of the WordNet and the RDFS version of classification system of Association for Computing Machinery Computing Classification System (ACM CCS); finally (5) Tests and validation of all developed functionalities. Findings and value The OmniPaper system can be instantiated to other domains other than news published in newspapers. The RSS technology is well suited for handling the description of scientific contents. RDF records that were used in the OmniPaper RDF prototype were replaced by RSS. The subject and lexical thesauri were kept. This strong metadata layer allows the creation of several services that facilitate the conceptual search of scientific contents. Originality and value of paper This paper presents a system that uses a central metadatabase to support conceptual searching mechanisms. The metadatabase consists of RDF triples generated from: (1) RSS files that were, by their turn generated from OAI-PMH harvested metadata records; (2) a controlled vocabulary (ACM-CCS) implemented in RDF Schema and (3) an RDF version of WordNet. This is a solution for a value-added service for the scientific community that is fully based in state-of-the-art standard technologies and is fully open for integration with other systems. Moreover this could be implemented by journals to improve the current mechanisms used to access, distribute and disseminate the scientific research developments.

Research limitations/implications (if applicable) The system implemented was tested but not evaluated in a real environment with specific users.