International Conference on Computer Systems and Technologies - CompSysTech’14 Multilingual Information Delivery Based on a Domain Ontology Nauris Paulins, Irina Arhipova, Signe Balina Abstract: As internationalisation becomes more and more intensive, organizations need to work with information in different languages in their work. Interlingua content delivery in e-learning courses could be effective learning content adaption method for delivering information for much wider audience. This paper suggests high-level framework for multilingual learning content delivery. This approach is based on a domain ontology, which allows to annotate and retrieve learning resources in a language independent way. Such method can improve accessibility, sharing and reuse of multilingual learning content. Key words: Adaptive learning, multilingual ontology, semantic annotation, domain adaption. INTRODUCTION The e-learning market has exploded over the last years. At the same time globalisation impact requests for multilingual content, where the same content must be produced in multiple languages. According to IEEE Learning Technology Standards Committee, a learning object (LO) is any entity, digital or non-digital, that may be used for learning [1]. E-learning courses are built for specific knowledge domain, but resources in these courses are divided into small chunks, that can be reused in various learning courses. But in different domains the meaning of a specialized term can differ, which is crucial for correct learning content translation. A specialized course content requires specialized knowledge on the domain, specific subject fields can always be approached from different perspectives, also they can change or develop in other directions, and translators need constant training in new subjects. That means that content should be adapted before it is presented to the user. Such process requires a great effort and huge amount of time, and as stated by Aguado and Montiel [2] translators need to repeat this training in all their working languages. As mentioned by several authors [3]–[5] the implementation of a semantic layer can improve the management of learning resources. Semantic layer basically is connected to the concept of ontologies. Ontology is metadata, which includes extra description about concepts and relationships of some specific field in course, therefore ontology is described as a specification of conceptualization [6]. An essential requirement for the metadata is their description in a machine readable way, thereby allowing interoperability on the Semantic Web [7]. Ontologies allow systems process data semantically, because there is a common knowledge base, made up of terms and the relationships between these terms. The analysis of literature has proved that the most acceptable form of ontologies for e-learning purpose is domain ontologies, which can be taxonomies [8] or more complex structure with several relations and rules [9]. The main factors to be considered for the usage of ontologies in e-learning systems are as follows: formalization level of ontologies and conceptualization of ontology. However, the multilingual information delivery does not include only content translation and annotation, but also such steps like information retrieval, storage and exchange. The translator can easily check the correspondence between terms in different language or find out variations between different conceptualizations. Researchers mostly analyse specific steps of this process, but do not include overall framework for multilingual information delivery. This paper proposes a framework for multilingual ontological knowledge delivery from content providers to students.