Type Prediction for Efficient Coreference Resolution in Heterogeneous Semantic Graphs
Loading...
Files
Links to Files
Permanent Link
Author/Creator
Author/Creator ORCID
Date
2013-09-16
Type of Work
Department
Program
Citation of Original Publication
Jennifer Sleeman and Tim Finin, Type Prediction for Efficient Coreference Resolution in Heterogeneous Semantic Graphs, 7th IEEE Int. Conf. on Semantic Computing, Sept. 2013, DOI: 10.1109/ICSC.2013.22
Rights
This item is likely protected under Title 17 of the U.S. Copyright Law. Unless on a Creative Commons license, for uses protected by Copyright Law, contact the copyright holder or the author.
© 2013 IEEE
© 2013 IEEE
Abstract
We describe an approach for performing entity type recognition in heterogeneous semantic graphs in order to reduce the computational cost of performing coreference resolution. Our research specifically addresses the problem of working with semi-structured text that uses ontologies that are not informative or not known. This problem is similar to coreference resolution in unstructured text, where entities and their types are identified using contextual information and linguistic-based analysis. Semantic graphs are semi-structured with very little contextual information and trivial grammars that do not convey additional information. In the absence of known ontologies, performing coreference resolution can be challenging. Our work uses a supervised machine learning algorithm and entity type dictionaries to map attributes to a common attribute space. We evaluated the approach in experiments using data from Wikipedia, Freebase and Arnetminer.