Probabilistic Parsing for German Using Sister-Head Dependencies

Probabilistic Parsing for German using Sister-Head Dependencies Amit Dubey Department of Computational Linguistics Saarland University PO Box 15 11 50 66041 Saarbr¨ucken, Germany adubey@coli.uni-sb.de Frank Keller School of Informatics University of Edinburgh 2 Buccleuch Place Edinburgh EH8 9LW, UK keller@inf.ed.ac.uk Abstract We present a probabilistic parsing model for German trained on the Negra tree- bank. We observe that existing lexicalized parsing models using head-head depen- dencies, while successful for English, fail to outperform an unlexicalized baseline model for German. Learning curves show that this effect is not due to lack of training data. We propose an alternative model that uses sister-head dependencies instead of head-head dependencies. This model out- performs the baseline, achieving a labeled precision and recall of up to 74%. This in- dicates that sister-head dependencies are more appropriate for treebanks with very flat structures such as Negra. 1 Introduction Treebank-based probabilistic parsing has been the subject of intensive research over the past few years, resulting in parsing models that achieve both broad coverage and high parsing accuracy (e.g., Collins 1997; Charniak 2000). However, most of the ex- isting models have been developed for English and trained on the Penn Treebank (Marcus et al., 1993), which raises the question whether these models generalize to other languages, and to annotation schemes that differ from the Penn Treebank markup. The present paper addresses this question by proposing a probabilistic parsing model trained on Negra (Skut et al., 1997), a syntactically annotated corpus for German. German has a number of syn- tactic properties that set it apart from English, and the Negra annotation scheme differs in important re- spects from the Penn Treebank markup. While Ne- gra has been used to build probabilistic chunkers (Becker and Frank, 2002; Skut and Brants, 1998), the research reported in this paper is the first attempt to develop a probabilistic full parsing model for Ger- man trained on a treebank (to our knowledge). Lexicalization can increase parsing performance dramatically for English (Carroll and Rooth, 1998; Charniak, 1997, 2000; Collins, 1997), and the lexi- calized model proposed by Collins (1997) has been successfully applied to Czech (Collins et al., 1999) and Chinese (Bikel and Chiang, 2000). However, the resulting performance is significantly lower than the performance of the same model for English (see Ta- ble 1). Neither Collins et al. (1999) nor Bikel and Chiang (2000) compare the lexicalized model to an unlexicalized baseline model, leaving open the pos- sibility that lexicalization is useful for English, but not for other languages. This paper is structured as follows. Section 2 re- views the syntactic properties of German, focusing on its semi-flexible wordorder. Section 3 describes two standard lexicalized models (Carroll and Rooth, 1998; Collins, 1997), as well as an unlexicalized baseline model. Section 4 presents a series of experi- ments that compare the parsing performance of these three models (and several variants) on Negra. The results show that both lexicalized models fail to out- perform the unlexicalized baseline. This is at odds with what has been reported …

Content truncated to 500 words