Vol.7 No.1 March 15, 2007       
Research Articles:
Ontology and Database Mapping: A Survey of Current Implementations  and Future Directions (pp001-024)
        N. Konstantinou, D.E. Spanos, and N. Mitrou
In this paper we discuss the problem of mapping relational database contents and ontologies. The motivation lies in the fact that during the latest years, the evolution in Web Technologies rendered the addition of intelligence to the information residing on the Web a necessity. We argue that the addition of formal semantics to the databases that store the majority of information found in the Web is important, in order to make this information searchable, accessible and retrievable. The key technologies towards this direction are the Semantic Web and the ontologies. We analyze in this paper the approaches that have so far been presented in order to exploit the prospects that such collaboration promises. We set the theoretical and practical boundaries of the mapping problem, we delve into the tools that altogether comprise today’s state of the art, and we provide a discussion about the benefits and the drawbacks of the existing approaches. We discuss the feasibility and viability of applying the mappings in real world applications as well as the directions that the evolution of current implementations should follow. We conclude by presenting the requirements that should be met in order to provide a more powerful next generation of mapping frameworks.

Reasoning on the Semantic Web for Adaptive Hypermedia  (pp025-041)
        L. Silva-Muñoz, K. Medina, M. Marsicano, M. Bonjour, and J. Palazzo
So far, ontologies have been widely used to convey knowledge across the Semantic Web. Complementing web ontologies with Horn-like rules to assert relations among ontology individuals and properties is part of the ongoing implementation of the Semantic Web. Intelligent Web Adaptive Hypermedia Systems \textit{(AHS)} are the next generation for adaptive hypermedia on the web. We present a web-based intelligent AHS for e-learning that configures on the fly complex learning objects tailored to the user profile. This automatic configuration is entirely accomplished by reasoning over a hybrid \textit{Knowledge Base (KB)} composed of ontologies, and Horn-like rules defined on top of ontologies concepts. Interoperability on the semantic level is achieved by using an {\it application profile} of standard vocabularies, standard languages for the representation of ontologies and rules, and a standard interface for reasoning functionality.

Boosting Computer Managed Instruction Functionalities Adoption in e-Learning Systems (pp042-069)
        G. Costagliola, F. Ferrucci, and V. Fuccella
Standardization efforts in e-learning are mainly aimed at achieving interoperability among Learning Management Systems (LMSs) and Learning Object (LO) authoring tools. In particular, the main standard producers are giving special attention to a set of functionalities, referred to as Computer Managed Instruction (CMI) and also known as SCORM Run-Time Environment.  Their adoption is crucial in the achievement of full interoperability among LMSs and LO authoring tools since they allow LOs to be launched in the LMS and to exchange data with it. Even desirable, standard compliancy and guideline adoption are difficult to obtain for LMS producers. This paper presents two design solutions aimed at boosting the adoption of CMI functionalities in Object-Oriented and Message-Oriented LMS systems, respectively. The former is a framework, named CMIFramework, which allows LMS developers to rapidly adopt CMI functionalities in Object-Oriented systems. The latter is a Service Oriented Architecture (SOA)-based reference model for offering the CMI functionalities as a service, external to the LMS. We investigate several case studies concerning the adoption of CMI functionalities, using our solutions, in different e-learning contexts.

Towards a Systematic Approach for the Credibility of Human-Centric Web Applications (pp070-092)
        P. Kamthan
The development and maintenance of Web Applications is viewed from an engineering perspective. A Pattern-Oriented Web Engineering Methodology (POWEM) for deploying patterns as means for improving the quality of Web Applications is presented. POWEM consists of a sequence of steps including the identification of stakeholder types, following a suitable development process model, identification of relevant quality attributes, and selection and application of suitable patterns. The feasibility issues involved in each step are examined. The use of patterns during macro- and micro-architecture design of a Web Application is illustrated. Finally, some directions for future research, including extensions to POWEM, are outlined.

Back to JWE Online Front Page