An Incremental Parser for Abstract Meaning Representation

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 536–546, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics An Incremental Parser for Abstract Meaning Representation Marco Damonte School of Informatics University of Edinburgh m.damonte@sms.ed.ac.uk Shay B. Cohen School of Informatics University of Edinburgh scohen@inf.ed.ac.uk Giorgio Satta Dept. of Information Engineering University of Padua satta@dei.unipd.it Abstract Abstract Meaning Representation (AMR) is a semantic representation for natural language that embeds annotations related to traditional tasks such as named entity recognition, semantic role labeling, word sense disambiguation and co-reference resolution. We describe a transition-based parser for AMR that parses sentences left- to-right, in linear time. We further pro- pose a test-suite that assesses specific sub- tasks that are helpful in comparing AMR parsers, and show that our parser is com- petitive with the state of the art on the LDC2015E86 dataset and that it outper- forms state-of-the-art parsers for recover- ing named entities and handling polarity. 1 Introduction Semantic parsing aims to solve the problem of canonicalizing language and representing its meaning: given an input sentence, it aims to ex- tract a semantic representation of that sentence. Abstract meaning representation (Banarescu et al., 2013), or AMR for short, allows us to do that with the inclusion of most of the shallow-semantic natural language processing (NLP) tasks that are usually addressed separately, such as named en- tity recognition, semantic role labeling and co- reference resolution. AMR is partially motivated by the need to provide the NLP community with a single dataset that includes basic disambiguation information, instead of having to rely on differ- ent datasets for each disambiguation problem. The annotation process is straightforward, enabling the development of large datasets. Alternative seman- tic representations have been developed and stud- ied, such as CCG (Steedman, 1996; Steedman, 2000) and UCCA (Abend and Rappoport, 2013). Several parsers for AMR have been recently de- veloped (Flanigan et al., 2014; Wang et al., 2015a; Peng et al., 2015; Pust et al., 2015; Goodman et al., 2016; Rao et al., 2015; Vanderwende et al., 2015; Artzi et al., 2015; Zhou et al., 2016). This line of research is new and current results suggest a large room for improvement. Greedy transition- based methods (Nivre, 2008) are one of the most popular choices for dependency parsing, because of their good balance between efficiency and ac- curacy. These methods seem promising also for AMR, due to the similarity between dependency trees and AMR structures, i.e., both representa- tions use graphs with nodes that have lexical con- tent and edges that represent linguistic relations. A transition system is an abstract machine char- acterized by a set of configurations and transitions between them. The basic components of a con- figuration are a stack of partially processed words and a buffer of unseen input words. Starting from an initial configuration, the system applies tran- sitions until a terminal configuration is reached. The sentence is scanned left to right, with linear time complexity …

Content truncated to 500 words