An ontology-based adaptive approach to P2P resource discovery in distributed scientific communities

Resource discovery is a challenge in a distributed environment. Trade-off is often needed between the speed and the accuracy of findings. There are two parts to resource discovery: the routing and matching of a query. This paper presents an adaptive approach to peer-to-peer (P2P) resource discovery which separates the routing of queries from the query matching mechanism. It focuses on improving the efficiency of routing search queries to increase the quality of the search results and also the scalability of the resource discovery in a highly decentralized P2P environment. This separation enables the adoption of any appropriate query matching/processing methods at a later stage. Three properties of scientific research communities provide the grounding for the approach: the existence of common interest groups, the willingness to share resources of common interests and the transitive relationship in the sharing behavior. The use of ontology enables ‘learning’ from past results and for providing guidance in future searches. By exploiting these features, the quality of search results can be improved and the network traffic reduced. Experimental results have provided some evidence to confirm the efficiency gain of this adaptive approach.