NYU's English ACE 2005 System Description
暂无分享,去创建一个
In this report we present the overall architecture for the NYU English ACE 2005 system. We focus on two components which were evaluated this year: reference resolution, where we experimented with features for relating anaphor and antecedent, and event recognition, where we sought to take advantage of a rich combination of logical grammatical structure and predicate-argument structure, building on the recent work on PropBank and NomBank. 1 System Overview Our English extraction system, developed over the course of the last several ACE evaluations, includes facilities for the EDR (entity), RDR (relation), and VDR (event) tasks; TIMEX and Value detection have not yet been implemented. Based on our most recent development work, we participated this year in two evaluations, for entities given mentions (to test coreference) and for events given entities, values, and time expressions. We present a complete system overview, although some components (such as ACE name recognition) were not used for the evaluations we performed this year.
[1] Ralph Grishman,et al. Parsing and GLARFing , 2001 .
[2] Ralph Grishman,et al. Extracting Relations with Integrated Information Using Kernel Methods , 2005, ACL.