MiRE4OWL

Abstract

Mobile user devices, such as smart phones have become the most popular highend users’ commodity. They will continue evolving toward being palm-top computers with the rapid development of wireless communications and platform technologies. Thanks to the improvement of the technology, the users’ interest and needs are heading for the realization of the Semantic Web in the mobile devices. To fulfill the needs basically, a semantic reasoner is required to understand the semantic knowledge formalism, namely ontology. The existing semantic reasoners designed for a mobile device was not practically enough to handle the OWL’s rich expressiveness due to lack of functionality. Moreover, after we evaluated the reasoners we noticed that they did not take care of the resource constraint of the mobile device well. In this thesis, we describe our effort to develop a practical semantic reasoner called MiRE4OWL to be embedded onto the resource-limited mobile user devices. MiRE4OWL is a rule-based inference engine which performs semantic reasoning operations over the application semantics represented in OWL-DL. It adopts a light-weight design to meet the resource constraints of the mobile devices and yet achieves better expressiveness than existing engines by diversifying logic operations, such as two types of negations and quantifiers. By applying MiRE4OWL to OWL test cases defined by the W3C, we got a plausible 83% success rate (i.e., pass 50 out of 60 cases). The performance evaluation showed that both the intended functionality and resource efficiency have been fulfilled.