A Semantics Extended Indexes framework for mashup discovery

Changbao Li,Bo Cheng,Junliang Junliang,Changjian Li,GongQiang Wang,Pingli Gu,Desheng Li
2011-01-01
Journal of Computational Information Systems
Abstract:The mashup is a hot research topic in recent years. In this paper, we propose a Semantics Extended Indexes (SEI) framework for mashup discovery. We design an SEI library creating algorithm to separate the semantics extension matching independence of the mashup discovery process, which improves the performance of the mashup discovery processing time largely. During SEI library creating, besides indexing on keywords of mashup profiles, we firstly make synonym extension on indexes based on synonym dictionary, and then make semantics ontology relation extension based on semantics ontology library. We apply the SEI framework in our mashup discovery system, and the discovery system performs high performance in both discovery processing time evaluation and precision/recall rate statistics. Copyright © 2011 Binary Information Press.
What problem does this paper attempt to address?