Anastasios Koutoumanos

PBURC: a patterns-based, unsupervised requirements clustering framework for distributed agile software development

Agile software development methodologies are increasingly adopted by organizations because they focus on the client’s needs, thus safeguarding business value for the final product. At the same time, as the economy and society move toward globalization, more organizations shift to distributed development of software projects. From this perspective, while adopting agile techniques seems beneficial, there are still a number of challenges that need to be addressed; among these notable is the effective cooperation between the stakeholders and the geographically distributed development team. In addition, data collection and validation for requirements engineering demands efficient processing techniques in order to handle the volume of data as well as to manage different inconsistencies, when the data are collected using online tools. In this paper, we present “PBURC,” a patterns-based, unsupervised requirements clustering framework, which makes use of machine-learning methods for requirements validation, being able to overcome data inconsistencies and effectively determine appropriate requirements clusters for optimal definition of software development sprints.

Applying metadata standards to multilingual learning objects

In this paper, we propose extensions to the IEEE Learning Object Metadata standard to provide the necessary mechanisms for describing and relating alternative language versions of learning resources. The work presented involves the identification of the various stakeholder applications and respective groups, the analysis of issues related to the metadescriptive needs of learning objects and their translations, the definition of an application profile capable of addressing these needs, and the compilation of per-case application guidelines. This proposal will be submitted to the IEEE Learning Technology Standardisation Committee, in order to be considered for the enhancement of a future version of the IEEE LOM standard.

From Concept to Sharing, to Delivery: Modeling a Quality Controlled Lifecycle for Learning Resources within a Federation of Repositories

An ever increasing number of members of educational communities engage in creating new digital resources within all fields of science and practice at all educational level within Learning Object Repositories, which, in terms, allows for even more efficient identification and delivery of existing material. This has in effect shaped a new type of activities related to the creation, sharing and re-use of learning objects through distributed repositories, associated within a federation with common policies, protocols and formats, which gave rise to an exponential growth of available material. However, this great success brings along a set of challenges, the most important being the demand for a quality controlled process, eliminating the information overload and safeguarding the added-value of learning object repositories, through models such as peer review and validation of learning objects and associated metadata. The authors of this paper provide insight on the approach for capitalizing on the European and international demand for quality-controlled educational content on Organic Agriculture and Agroecology, describing a framework and the associated quality issues for modeling the publication lifecycle of digital learning objects within a federation of learning object repositories.

Ontology-driven linked open (meta-) data services for e-research systems

This paper presents an ontology-driven approach for implementing Linked Open Data services for scholarly and research artifacts, to facilitate the dissemination of research results. The resulting ontology and the associated metadata Application Profile, built upon the Singapore Framework for Dublin Core Application Profiles, have been used to enable the development of interoperable systems and services for the research community in the agriculture and aquaculture domains. The paper describes the implementation approach for the AgRes ontology, the ontology itself as well as the interoperability levels of the AgRes application profile. Furthermore, it outlines the creation and the management of the ontology with the MoKi collaborative Media Wiki-based tool.

Multilingual Educational Repositories: Breaking Down the Language Barrier

This paper presents a framework for introducing multilingualism into web portals, aiming to facilitate bringing down the language barriers, bridging the language divide between information seekers and the wealth of resources from around the world that are made available through such portals. The main research question addressed is the definition of the problem itself and the description of design affordances that shall allow web portals to meet different linguistic profiles, needs and expectations from users around the globe. The paper describes the methodology used for the analysis of user requirements of the international user base of a thematic web portal aiming to promote discovery of educational resources and concludes with some initial results, discussion of the findings and suggestions for future work.

Design for all: towards a social platform for integrating distributed open-access repositories

Recent advances in networking and telecommunications technologies combined with the vast load of readily available scientific data, urge towards the implementation of software systems that are able to collect, combine and present information from distributed repositories. At the same time, scientists, researchers, academics and experts demand flexible infrastructures that provide just-in-time information, with reliable, peer-reviewed data, through ubiquitous interfaces. Towards this direction, several international organizations including the European Commission, promote the development of software platforms that will facilitate universal access to information aggregated from open access repositories. In this paper we discuss the requirements engineering process and the key functionality for a system that supports knowledge extraction and dissemination from distributed repositories, while at the same time enabling and integrating the social interaction of its users.