2007 UnsupervisedCorefRes

From GM-RKB
Jump to navigation Jump to search

Subject Headings: Unsupervised Mention Coreference Resolution Algorithm, Unsupervised Coreference Resolution System.

Notes

  • (Wick et al., 2009) ⇒ Michael Wick, Aron Culotta, Khashayar Rohanimanesh, and Andrew McCallum. (2009). “An Entity Based Model for Coreference Resolution.” In: Proceedings of the SIAM International Conference on Data Mining (SDM 2009).
    • QUOTE: Haghighi and Klein [23] propose an unsupervised Bayesian model for newswire coreference. In this generative model, each mention is drawn from a latent entity. However, since each attribute is a distribution over words, the model does not produce a single canonical representation for each entity, a vital step that would have to be performed post-hoc to store the entities in a first normal form relational database. In contrast our system produces canonical representations automatically as coreference is performed. Also, our model is discriminatively trained allowing it to capture arbitrary dependencies between features without the addition of extra edges in the graphical sense.

Cited By

Quotes

Abstract

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.

References


,

 AuthorvolumeDate ValuetitletypejournaltitleUrldoinoteyear
2007 UnsupervisedCorefResDan Klein
Aria Haghighi
Unsupervised Coreference Resolution in a Nonparametric Bayesian Modelhttp://acl.ldc.upenn.edu/P/P07/P07-1107.pdf