A Transition-Based Directed Acyclic Graph Parser for UCCA

Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1127–1138 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1104 Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 1127–1138 Vancouver, Canada, July 30 - August 4, 2017. c⃝2017 Association for Computational Linguistics https://doi.org/10.18653/v1/P17-1104 A Transition-Based Directed Acyclic Graph Parser for UCCA Daniel Hershcovich1,2 Omri Abend2 1The Edmond and Lily Safra Center for Brain Sciences 2School of Computer Science and Engineering Hebrew University of Jerusalem {danielh,oabend,arir}@cs.huji.ac.il Ari Rappoport2 Abstract We present the first parser for UCCA, a cross-linguistically applicable framework for semantic representation, which builds on extensive typological work and sup- ports rapid annotation. UCCA poses a challenge for existing parsing techniques, as it exhibits reentrancy (resulting in DAG structures), discontinuous structures and non-terminal nodes corresponding to com- plex semantic units. To our knowledge, the conjunction of these formal properties is not supported by any existing parser. Our transition-based parser, which uses a novel transition set and features based on bidirectional LSTMs, has value not just for UCCA parsing: its ability to handle more general graph structures can inform the de- velopment of parsers for other semantic DAG structures, and in languages that fre- quently use discontinuous structures. 1 Introduction Universal Conceptual Cognitive Annotation (UCCA, Abend and Rappoport, 2013) is a cross- linguistically applicable semantic representation scheme, building on the established Basic Lin- guistic Theory typological framework (Dixon, 2010a,b, 2012), and Cognitive Linguistics litera- ture (Croft and Cruse, 2004). It has demonstrated applicability to multiple languages, including English, French, German and Czech, support for rapid annotation by non-experts (assisted by an accessible annotation interface (Abend et al., 2017)), and stability under translation (Sulem et al., 2015). It has also proven useful for machine translation evaluation (Birch et al., 2016). UCCA differs from syntactic schemes in terms of content and formal structure. It exhibits reentrancy, discontinuous nodes and non-terminals, which no single existing parser supports. Lacking a parser, UCCA’s applicability has been so far limited, a gap this work addresses. We present the first UCCA parser, TUPA (Transition-based UCCA Parser), building on re- cent advances in discontinuous constituency and dependency graph parsing, and further introduc- ing novel transitions and features for UCCA. Transition-based techniques are a natural starting point for UCCA parsing, given the conceptual similarity of UCCA’s distinctions, centered around predicate-argument structures, to distinctions ex- pressed by dependency schemes, and the achieve- ments of transition-based methods in dependency parsing (Dyer et al., 2015; Andor et al., 2016; Kiperwasser and Goldberg, 2016). We are fur- ther motivated by the strength of transition-based methods in related tasks, including dependency graph parsing (Sagae and Tsujii, 2008; Ribeyre et al., 2014; Tokg¨oz and Eryi˘git, 2015), con- stituency parsing (Sagae and Lavie, 2005; Zhang and Clark, 2009; Zhu et al., 2013; Maier, 2015; Maier and Lichte, 2016), AMR parsing (Wang et al., 2015a,b, 2016; Misra and Artzi, 2016; Goodman et al., 2016; Zhou et al., 2016; Damonte et al., 2017) and CCG parsing …

Content truncated to 500 words