Modeling RDF Data for MetOcean Information Systems Kamaluddeen Usman Danyaro and Jafreezal Jaafar Department of Computer & Information Sciences, Universiti Teknologi PETRONAS, Tronoh Perak, Malaysia Email: kamaluddeen_g01325@utp.com.my, jafreez@petronas.com.my M. S. Liew Department of Civil Engineering, Universiti Teknologi PETRONAS, Tronoh Perak, Malaysia Email: shahir_liew@petronas.com.my Abstract—This paper suggests a Resource Description Framework (RDF) model for effective handling of distributed data relating to the Semantic Web. RDF is a data model that integrates aptly structured data. The World Wide Web is facing great challenges in data retrieval as a result of increase and request of knowledge from different sources. This brings the issue of information overload. Therefore, in minimizing this challenge, the paper presents the process of modeling RDF knowledge base which is the essential step of Semantic Web development. As a state-of- the-art research, a meteorological and oceanographic dataset has been used to provide the basic concepts of RDF serialization. The result accessed through the developed SPARQL endpoint. Thus, it implies that the approach is sufficient for querying as well as data representation of distributed data. Index Terms—Resource Description Framework (RDF), RDF Model, Semantic Web, MetOcean I. INTRODUCTION Since from the inception of Resource Description Framework (RDF) in 1990s, various processes have been followed as a way to produce the ideal model or structure for language specification. In 2004, there were extended RDF specifications for representing semantic data that have been standardized by World Wide Web Consortium (W3C). RDF is the most essential source of Semantic Web that works with various things on the networks. It manages and handles the distributed data which gives techniques for data representation [1]. The representation of data from unstructured information into organized information are called modeling [1]. Nevertheless, RDF is a language for specifying languages. It is among the three modeling languages (RDF, RDFs and Web Ontology Language) of Semantic Web. These languages stipulate the expressivity and structure within the Web knowledge. Semantic Web is about information systems integration in meaningful way. Therefore, it depends on information sharing between community as well as the systems. An RDF application follows the typical sources of data. This can be done by integrating the data with relational database management system and later incorporate it with the Web through Uniform Resource Identifiers (URIs). URI also offers a good level of data presentation and enabling shared data from different sources; global references. The infrastructure is laid over the distributed networks through Uniform Resource Locator (URL). Then become interoperable by exchanging and handling the eXtensible Markup Language (XML) data [2, 3]. XML has been the mediator between human and computers (on the Web) with the help of standard syntax that produce documents. Although, XML and database supply the consistency of every Web that initiate all sorts of data models [1]. However, XML as being a data model alone cannot handle large amount of data and system interoperability. It is because the XML provides syntax not semantics for data representation over the web. Nevertheless, many Web database systems are not on Linked Data model structure. For instance, the database of Meteorological and Oceanographic (MetOcean) information system is built on XML and XML metadata only [4]. MetOcean is a meteorological and oceanographic industry that handles large amounts of data and metadata. A lot of companies and research industries depend on its data that has been built on XML schema. However, these tight on the capabilities of describing data or metadata in meaningful which leads to the problems of information overload. Information overload is the process of getting information that might not be sufficiently organized as a result of rapid advancement of information and communication technology [5-7]. The aim of this paper is to define and describe the distributed data of MetOcean in a meaningful way. In this regard, we design a network graph, perhaps a triplestore that stores graphs of MetOcean’s semantic repository and able to be queried as linked data. The structure of this paper has the following outline. Section 2 provides the RDF and its basic concepts. Section 3 discusses the need for RDF model in meteorological and oceanographic information systems. Section 4 provides the related works. In section 5 we present the proposed RDF data model consisting of the structure of MetOcean’s triplestore, RDF representation 432 JOURNAL OF COMPUTERS, VOL. 9, NO. 2, FEBRUARY 2014 © 2014 ACADEMY PUBLISHER doi:10.4304/jcp.9.2.432-440