Paper: Classifier Combination Techniques Applied to Coreference Resolution

ACL ID N09-3001
Title Classifier Combination Techniques Applied to Coreference Resolution
Venue HLT-NAACL Companion Volume: Student Research Workshop and Doctoral Consortium
Session
Year 2009
Authors

This paper examines the applicability of clas- sifier combination approaches such as bagging and boosting for coreference resolution. To the best of our knowledge, this is the first ef- fort that utilizes such techniques for corefer- ence resolution. In this paper, we provide ex- perimental evidence which indicates that the accuracy of the coreference engine can po- tentially be increased by use of bagging and boosting methods, without any additional fea- tures or training data. We implement and eval- uate combination techniques at the mention, entity and document level, and also address is- sues like entity alignment, that are specific to coreference resolution.