Automated ontology population and enrichment of scientific publications

Scientific publications are the most important resources available to the research communities. Researchers want their work to be widely recognized and available and also need powerful search engines to identify other publications and researchers working in the same area. Therefore, a good representation and organization of scientific products is crucial for an accurate retrieval of information. This paper describes an approach for automated population and semantic enrichment of an ontology model that represents scientific publications. Specifically, the type of enrichment used in this approach consists of implementing semantic similarity measurements between publications. Several experiments were performed to identify the best similarity measurement, using a statistical approach and the precision of the measurements