Employing Ontologies In Resource-scarce Mobile Ad-hoc Networks : Translating From The Web Ontology Language (OWL) To Topic Maps

Abstract

In a rescue scenario, you have personnel from different organizations cooperating, and this personnel has to communicate both within their own organization and with personnel from other organizations. If they are carrying handheld devices, they can receive and send automatic information updates within a mobile ad-hoc network that consists of all of the handheld devices carried by rescue personnel and sensors that are within range. A challenge with this kind of information sharing is that different organizations may use different data models and vocabularies for defining the same concepts. Ontologies can be used as a bridge between these different vocabularies, enabling a mapping between the concepts of the different vocabularies. The Web Ontology Language (OWL) can be used to express the ontologies. OWL is a very expressive language, and has the advantage that it is possible to perform reasoning over ontologies and infer knowledge that is not explicitly stated. The only problem is that reasoning engines for OWL typically require a lot of resources, and are therefore not well suited for resource-limited handheld devices. Topic maps is another technology for expressing ontologies. Topic maps are less expressive than OWL and do not provide support for automated reasoning, but there exists a topic map engine that allows you to browse and query the topic map and that can be used on resourcelimited devices. OWL is built on another language called the Resource Description Framework (RDF), and the topic map and RDF communities have looked at how these two different languages can be translated into each other. Since OWL is built on RDF, we look into if any of the translation proposals for translating between RDF and topic maps also can be used for translating between OWL and topic maps. This will allow us to create an ontology in OWL, perform reasoning on the ontology and add the information inferred from the reasoning to the ontology directly. Then the ontology can be translated into a topic map, and can be used on a handheld device. This thesis looks at how one can translate ontologies from OWL to topic maps, and how usable the resulting topic maps are in a mobile ad-hoc network for a rescue operation.