Paper: Combining Compositionality and Pagerank for the Identification of Semantic Relations between Biomedical Words

ACL ID W12-2413
Title Combining Compositionality and Pagerank for the Identification of Semantic Relations between Biomedical Words
Venue Workshop on Biomedical Natural Language Processing
Session
Year 2012
Authors

The acquisition of semantic resources and re- lations is an important task for several appli- cations, such as query expansion, information retrieval and extraction, machine translation. However, their validity should also be com- puted and indicated, especially for automatic systems and applications. We exploit the com- positionality based methods for the acquisi- tion of synonymy relations and of indicators of these synonyms. We then apply pager- ank-derived algorithm to the obtained seman- tic graph in order to filter out the acquired syn- onyms. Evaluation performed with two inde- pendent experts indicates that the quality of synonyms is systematically improved by 10 to 15% after their filtering.