Parsing Universal Dependencies without training

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 230–240, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics Parsing Universal Dependencies without training Héctor Martínez Alonso♠ Željko Agi´c♥ Barbara Plank♣ Anders Søgaard♦ ♠Univ. Paris Diderot, Sorbonne Paris Cité – Alpage, INRIA, France ♥IT University of Copenhagen, Denmark ♣Center for Language and Cognition, University of Groningen, The Netherlands ♦University of Copenhagen, Denmark hector.martinez-alonso@inria.fr,zeag@itu.dk,b.plank@rug.nl,soegaard@di.ku.dk Abstract We propose UDP, the first training-free parser for Universal Dependencies (UD). Our algorithm is based on PageRank and a small set of head attachment rules. It fea- tures two-step decoding to guarantee that function words are attached as leaf nodes. The parser requires no training, and it is competitive with a delexicalized transfer system. UDP offers a linguistically sound unsupervised alternative to cross-lingual parsing for UD, which can be used as a baseline for such systems. The parser has very few parameters and is distinctly ro- bust to domain change across languages. 1 Introduction Grammar induction and unsupervised dependency parsing are active fields of research in natural language processing (Klein and Manning, 2004; Gelling et al., 2012). However, many data-driven approaches struggle with learning relations that match the conventions of the test data, e.g., Klein and Manning reported the tendency of their DMV parser to make determiners the heads of German nouns, which would not be an error if the test data used a DP analysis (Abney, 1987). Even super- vised transfer approaches (McDonald et al., 2011) suffer from target adaptation problems when fac- ing word order differences. The Universal Dependencies (UD) project (Nivre et al., 2015; Nivre et al., 2016) offers a dependency formalism that aims at providing a consistent representation across languages, while enforcing a few hard constraints. The arrival of such treebanks, expanded and improved on a reg- ular basis, provides a new milestone for cross- lingual dependency parsing research (McDonald et al., 2013). Furthermore, given that UD rests on a series of simple principles like the primacy of lexical heads, cf. Johannsen et al. (2015) for more details, we expect that such a formalism lends itself more nat- urally to a simple and linguistically sound rule- based approach to cross-lingual parsing. In this paper we present such an approach. Our system is a dependency parser that requires no training, and relies solely on explicit part-of- speech (POS) constraints that UD imposes. In par- ticular, UD prescribes that trees are single-rooted, and that function words like adpositions, auxil- iaries, and determiners are always dependents of content words, while other formalisms might treat them as heads (De Marneffe et al., 2014). We as- cribe our work to the viewpoints of Bender (2009) about the incorporation of linguistic knowledge in language-independent systems. Contributions We introduce, to the best of our knowledge, the first unsupervised rule-based de- pendency parser for Universal Dependencies. Our method goes substantially beyond the exist- ing work on rule-aided unsupervised dependency parsing, specifically by: i) adapting the dependency head rules …

Content truncated to 500 words