SERSE: searching for semantic web content
Valentina A. M. Tamma, Ian Blacoe, Ben Lithgow Smith, Michael Wooldridge · 2004
Abstract. Searching the Semantic Web (SW) for digital content is arguably more complex than searching the current web. In fact, a SW search system must deal with a large number of distributed, heterogeneous resources, that may reference many different ontologies. In order to manage this complexity we have integrated different technologies such as peer-to-peer, ontologies, and multi-agent technology in the design of SERSE, a multi-agent system for searching the SW. In SERSE, agents are organised in a P2P network according to a semantic overlay, where the neighbourhood is determined by the semantic proximity of the ontological definitions that are known to the agents. The integration of these technologies poses some problems. On the one hand, the more ontological knowledge the agents have, the better we can expect the system to perform. On the other hand, global knowledge would constitute a point of centralisation which might potentially degrade the performance of a P2P system. In addition, maintenance of such semantic overlay can also degrade the performance. The paper presents SERSE together with some experimental results that evaluate the performance in response to changes in the size of semantic neighbourhood and an analytical evaluation of the workload of the system due to maintenance activities. 1