Collaborative Partitioning for Coreference Resolution

Abstract

This paper presents a collaborative partitioning algorithm-a novel ensemblebased approach to coreference resolution. Starting from the all-singleton partition, we search for a solution close to the ensemble's outputs in terms of a task-specific similarity measure. Our approach assumes a loose integration of individual components of the ensemble and can therefore combine arbitrary coreference resolvers, regardless of their models. Our experiments on the CoNLL dataset show that collaborative partitioning yields results superior to those attained by the individual components, for ensembles of both strong and weak systems. Moreover, by applying the collaborative partitioning algorithm on top of three state-of-the-art resolvers, we obtain the second-best coreference performance reported so far in the literature (MELA v08 score of 64.47).

Improving mention detection for Basque based on a deep error analysis

“…Given the above, we can conclude that the input that our mention detector receives has a substantial impact on mention detection errors. This is also pointed out by Uryupina and Moschitti (2013), who suggest that better performance can be obtained if a robust preprocessing is achieved. The performance of the mention detector is highly dependent on the preprocessing tools, and errors caused by these tools are difficult to tackle.…”

“…In end-to-end coreference resolution systems, low recall in the mention detection step has negative effects on the whole process, since missed mentions are not available for coreference resolution in the later stages. As is pointed out in Uryupina and Moschitti (2013), in the CoNLL 2011 shared task the majority of the participants relied on rule-based modules to obtain mention boundaries for English. However, in the CoNLL 2012 shared task, participants with rule-based systems fell back to very simple baselines in mention detection, which demonstrated that rule-based systems are not easily adaptable to languages other than the one for which they were created.…”