This is a past event
Abstract: While more and more RDF data is available, the question of how typical web users can access this body of knowledge becomes of crucial importance. There is thus a growing amount of research on natural language interfaces that allow end users to profit from the expressive power of Semantic Web standards while hiding their complexity behind an intuitive and easy-to-use interface.
The key challenge for question answering over RDF data is to translate the users' information need into a form such that it can be evaluated using standard Semantic Web query processing and inferencing techniques. I will give a brief overview of the main challenges involved in getting from natural language questions to structured queries and answers, and present a template-based approach to addressing these challenges, that focuses on the integration of deep semantic interpretation and schema-agnostic, statistical methods.
Bio: Christina Unger is a postdoctoral researcher in the Semantic Computing group affiliated to the Cluster of Excellence on Cognitive Interaction Technology (CITEC) at Bielefeld University, Germany. She received her PhD from the Utrecht Institute of Linguistics (UiL-OTS), the Netherlands, in 2010, in the area of computational semantics. Her major research interest are natural language interfaces to structured data, with a focus on ontology-based natural language understanding, question answering over RDF data, and the automatic induction of grammars from the lexicon-ontology interface.
Links:
Paper "Template-based question answering over linked data": http://dl.acm.org/citation.cfm?id=2187923
OKBQA hackathon and open platform: http://www.okbqa.org/
Benchmarking campaign: http://www.sc.cit-ec.uni-bielefeld.de/qald/
- Speaker
- Christina Unger
- Hosted by
- Adam Wyner
- Venue
- MT 203