Paolo Cifariello, Paolo Ferragina and Marco Ponza from the University of Pisa, are presenting a new semantic search engine for expert finding in the academia called WISER. In their paper titled, WISER: A Semantic Approach for Expert Finding in Academia based on Entity Linking a system, the authors note that their system is unsupervised and jointly combines classical language modeling techniques based on text evidence with the Wikipedia knowledge graph via entity linking.
WISER, is a novel search engine for finding experts in the academia. The novelty relies on the deployment of entity linking in addition to relatedness and entity embedding. The creation of a web experience management (WEM) profile is based on a weighted and labeled graph, which models the explicit knowledge of an author by means of explicit and latent concepts occurring in their documents and semantic relations.
Experiments have shown that ranking authors according to the semantic relation between the user query and the WEM profile achieves state-of-the-art performance, thus making WISER the best publicly available software for academic-expert finding.
The effectiveness of the system was established over a large-scale test on a standard data set for this task. The test revealed that WISER outperforms its competitors and proved the effectiveness of modeling author profiles via semantic graph of entities.
Click here to read the complete paper.
Please give your feedback on this article or share a similar story for publishing by clicking here.
