A mixed approach for the representation of nutritional information through XML-to-OWL mappings

Abstract

Semantic Web technologies (SWTs), such as XML and OWL ontologies are increasingly being used to represent information in different domains. However, these capabilities are not indistinctly provided by each SWT. XML, although not being considered a SWT in itself, stands at the syntactic level of the Semantic Web stack, and is more suitable for efficient information structure and retrieval in interactive software applications. OWL language, on the other hand, is more suitable for background reasoning and consistency checking purposes. In this paper, we introduce a mixed approach for the information representation and knowledge sharing in the nutritional domain, aiming to explode XML and OWL benefits. This approach is included in NutElCare (a nutritional recommender system). In it, diets are represented through XML documents contained in an XML repository and the knowledge base is composed of several OWL ontologies which interact to provide recommendations. In this design, XSLT transformations play an important role, allowing the mappings from XML diets to the OWL ontologies, so that elevating the syntactic representation of the XML documents to the semantic level of OWL ontologies. Altogether, they configure a system architecture that keeps the system timely responsive through a seamless linkage between XML and OWL representations.