Simple coreference resolution with rich syntactic and semantic features

Abstract

Coreference systems are driven by syntactic, semantic, 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 discourse component, our system is deterministic and is driven entirely by syntactic and semantic compatibility 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 contributions include (1) the presentation of a simpleto-reproduce, high-performing baseline and (2) the demonstration that most remaining errors can be attributed to syntactic and semantic factors external to the coreference phenomenon (and perhaps best addressed by non-coreference systems).

A New Entity Salience Task with Millions of Training Examples

“…This includes a POS tagger and dependency parser, comparable in accuracy to the current Stanford dependency parser (Klein and Manning, 2003); an NP extractor that uses POS tags and dependency edges to identify a set of entity mentions; a coreference resolver, comparable to that of Haghighi and Klein, (2009) for clustering mentions; and an entity resolver that links entities to Freebase profiles. The entity resolver is described in detail by Lao, et al (2012).…”

Le corpus ANCOR_Centre et son outil de requêtage : application à l’étude de l’accord en genre et nombre dans les coréférences et anaphores en français parlé

“…Les résultats présentés dans ce paragraphe les concernent donc directement. (Haghighi & Klein, 2009). Dans le cas des systèmes centrés données reposant sur un apprentissage sur corpus annoté, on remarquera que des propositions ont été faites pour apprendre des modèles différents suivant le type d'entité considéré (Ng, 2005).…”

“…De son côté, l'importance de la résolution des coréférences et anaphores a été reconnue par le TALN dès les travaux pionniers de Hobbs (1978). Aux travaux initiaux basés sur des méthodes symboliques (Lappin et Leas, 1994) ont succédé des approches plus heuristiques (Mitkov, 1994), avant que les approches par apprentissage sur données (Soon et al, 2001, Ng et Cardie, 2002, Haghighi & Klein 2009) ne deviennent prédominantes.…”

Modelling Events through Memory-based, Open-IE Patterns for Abstractive Summarization

“…The news have been processed with a tokenizer, a sentence splitter (Gillick and Favre, 2009), a part-of-speech tagger and dependency parser (Nivre, 2006), a co-reference resolution module (Haghighi and Klein, 2009) and an entity linker based on Wikipedia and Freebase (Milne and Witten, 2008). We use Freebase types as finegrained named entity types, so we are also able to label e.g.…”