Statistical Parsing for Information Extraction from Scientific Articles
This event took place on Wednesday 17 November 2004 at 12:45
Dr. Ted Briscoe University of Cambridge Computer Laboratory
I'll describe the states-of-the-art in statistical parsing and information extraction (IE), present the RASP (Robust Accurate Statistical Parsing) System for English, and outline our project with FlyBase, Cambridge to develop an IE system capable of supporting efficient curation of functional genomic information from the fruit fly literature.
I'll argue that IE from the full text of scientific articles requires full statistical parsing, both to reliably locate useful information and to integrate the IE system effectively with domain resources, and that migration to the semantic web/grid creates exciting opportunities, both for generic integration of IE into curation and experimentation and efficient porting of IE to new domains.
This event took place on Wednesday 17 November 2004 at 12:45
Dr. Ted Briscoe University of Cambridge Computer Laboratory
I'll describe the states-of-the-art in statistical parsing and information extraction (IE), present the RASP (Robust Accurate Statistical Parsing) System for English, and outline our project with FlyBase, Cambridge to develop an IE system capable of supporting efficient curation of functional genomic information from the fruit fly literature.
I'll argue that IE from the full text of scientific articles requires full statistical parsing, both to reliably locate useful information and to integrate the IE system effectively with domain resources, and that migration to the semantic web/grid creates exciting opportunities, both for generic integration of IE into curation and experimentation and efficient porting of IE to new domains.
