Joint Dependency Parsing and Multiword Expression Tokenization

Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing, pages 1116–1126, Beijing, China, July 26-31, 2015. c⃝2015 Association for Computational Linguistics Joint Dependency Parsing and Multiword Expression Tokenisation Alexis Nasr, Carlos Ramisch, Jos´e Deulofeu, Andr´e Valli Aix Marseille Universit´e, CNRS, LIF UMR 7279 Marseille, France FirstName.LastName@lif.univ-mrs.fr Abstract Complex conjunctions and determiners are often considered as pretokenized units in parsing. This is not always realistic, since they can be ambiguous. We pro- pose a model for joint dependency parsing and multiword expressions identification, in which complex function words are rep- resented as individual tokens linked with morphological dependencies. Our graph- based parser includes standard second- order features and verbal subcategoriza- tion features derived from a syntactic lex- icon.We train it on a modified version of the French Treebank enriched with mor- phological dependencies. It recognizes 81.79% of ADV+que conjunctions with 91.57% precision, and 82.74% of de+DET determiners with 86.70% precision. 1 Introduction Standard NLP tool suites for text analysis are of- ten made of several processes that are organized as a pipeline, in which the input of a process is the output of the preceding one. Among these processes, one commonly finds a tokenizer, which segments a sentence into words, a part-of-speech (POS) tagger, which associates to every word a part-of-speech tag, and a syntactic parser, which builds a parse tree for the sentence1. These three processes correspond to three formal operations on the string: segmentation into linguistically rel- evant units (words), tagging the words with POS tags and linking the (word, POS) pairs by means of syntactic dependencies. This setup is clearly not ideal, as some decisions are made too early in the pipeline (Branco and Silva, 2003). More specifically, some tokenization and tagging choices are difficult to make without 1This paper considers dependency syntactic structures. taking syntax into account. To avoid the pitfall of premature decisions, probabilistic tokenizers and taggers can produce several solutions in the form of lattices (Green and Manning, 2010; Goldberg and Elhadad, 2011). Such approaches usually lead to severe computational overhead due to the huge search space in which the parser looks for the opti- mal parse tree. Besides, the parser might be biased towards short solutions, as it compares scores of trees associated to sequences of different lengths (De La Clergerie, 2013). This problem is particularly hard when parsing multiword expressions (MWEs), that is, groups of tokens that must be treated as single units (Bald- win and Kim, 2010). The solution we present in this paper is different from the usual pipeline. We propose to jointly parse and tokenize MWEs, transforming segmentation decisions into linking decisions. Our experiments concentrate on two difficult tokenization cases. Hence, it is the parser that will choose, in such cases, whether to group or not several tokens. Our first target phenomenon is the family of ADV+que constructions, a type of complex con- junction in French. They are formed by adverbs like bien (well) or ainsi (likewise) …

Content truncated to 500 words