Unsupervised Coreference Resolution in a Nonparametric Bayesian Model

We present an unsupervised, nonparametric Bayesian approach to coreference resolution which models both global entity identity across a corpus as well as the sequential anaphoric structure within each document. While most existing coreference work is driven by pairwise decisions, our model is fully generative, producing each mention from a combination of global entity properties and local attentional state. Despite being unsupervised, our system achieves a 70.3 MUC F1 measure on the MUC-6 test set, broadly in the range of some recent supervised results.

• Type: Conference
• Year: 2007
• Where: ACL
• Authors: Aria Haghighi, Dan Klein
• Keywords: ACL 2007, Computational Linguistics, Global Entity, Global Entity Identity, Nonparametric Bayesian Approach