Simple Coreference Resolution with Rich Syntactic and Semantic Features

Proceedings of the 2009 Conference on Empirical Methods in Natural Language Processing, pages 1152–1161, Singapore, 6-7 August 2009. c⃝2009 ACL and AFNLP Simple Coreference Resolution with Rich Syntactic and Semantic Features Aria Haghighi and Dan Klein Computer Science Division UC Berkeley {aria42, klein}@cs.berkeley.edu Abstract Coreference systems are driven by syntactic, se- mantic, and discourse constraints. We present a simple approach which completely modularizes these three aspects. In contrast to much current work, which focuses on learning and on the dis- course component, our system is deterministic and is driven entirely by syntactic and semantic com- patibility as learned from a large, unlabeled corpus. Despite its simplicity and discourse naivete, our system substantially outperforms all unsupervised systems and most supervised ones. Primary con- tributions include (1) the presentation of a simple- to-reproduce, high-performing baseline and (2) the demonstration that most remaining errors can be at- tributed to syntactic and semantic factors external to the coreference phenomenon (and perhaps best addressed by non-coreference systems). 1 Introduction The resolution of entity reference is influenced by a variety of constraints. Syntactic constraints like the binding theory, the i-within-i filter, and appos- itive constructions restrict reference by configura- tion. Semantic constraints like selectional compat- ibility (e.g. a spokesperson can announce things) and subsumption (e.g. Microsoft is a company) rule out many possible referents. Finally, dis- course phenomena such as salience and centering theory are assumed to heavily influence reference preferences. As these varied factors have given rise to a multitude of weak features, recent work has focused on how best to learn to combine them using models over reference structures (Culotta et al., 2007; Denis and Baldridge, 2007; Klenner and Ailloud, 2007). In this work, we break from the standard view. Instead, we consider a vastly more modular system in which coreference is predicted from a determin- istic function of a few rich features. In particu- lar, we assume a three-step process. First, a self- contained syntactic module carefully represents syntactic structures using an augmented parser and extracts syntactic paths from mentions to potential antecedents. Some of these paths can be ruled in or out by deterministic but conservative syntactic constraints. Importantly, the bulk of the work in the syntactic module is in making sure the parses are correctly constructed and used, and this mod- ule’s most important training data is a treebank. Second, a self-contained semantic module evalu- ates the semantic compatibility of headwords and individual names. These decisions are made from compatibility lists extracted from unlabeled data sources such as newswire and web data. Finally, of the antecedents which remain after rich syntac- tic and semantic filtering, reference is chosen to minimize tree distance. This procedure is trivial where most systems are rich, and so does not need any supervised corefer- ence data. However, it is rich in important ways which we argue are marginalized in recent coref- erence work. Interestingly, error analysis from our final system shows that its failures are far more often due to syntactic failures (e.g. parsing mis- takes) …

Content truncated to 500 words