Experiments with Easy-first nonprojective constituent parsing
نویسنده
چکیده
Less-configurational languages such as German often show not just morphological variation but also free word order and nonprojectivity. German is not exceptional in this regard, as other morphologically-rich languages such as Czech, Tamil or Greek, offer similar challenges that make context-free constituent parsing less attractive. Advocates of dependency parsing have long pointed out that the free(r) word order and nonprojective phenomena are handled in a more straightforward way by dependency parsing. However, certain other phenomena in language, such as gapping, ellipses or verbless sentences, are difficult to handle in a dependency formalism. In this paper, we show that parsing of discontinuous constituents can be achieved using easy-first parsing with online reordering, an approach that previously has only been used for dependencies, and that the approach yields very fast parsing with reasonably accurate results that are close to the state of the art, surpassing existing results that use treebank grammars. We also investigate the question whether phenomena where dependency representations may be problematic – in particular, verbless clauses – can be handled by this model.
منابع مشابه
Probabilistic Models of Nonprojective Dependency Trees
A notable gap in research on statistical dependency parsing is a proper conditional probability distribution over nonprojective dependency trees for a given sentence. We exploit the Matrix Tree Theorem (Tutte, 1984) to derive an algorithm that efficiently sums the scores of all nonprojective trees in a sentence, permitting the definition of a conditional log-linear model over trees. While discr...
متن کاملIncorporating Semi-supervised Features into Discontinuous Easy-First Constituent Parsing
This paper describes adaptations for EAFI, a parser for easy-first parsing of discontinuous constituents, to adapt it to multiple languages as well as make use of the unlabeled data that was provided as part of the SPMRL shared task 2014.
متن کاملPseudo-Projective Dependency Parsing
In order to realize the full potential of dependency-based syntactic parsing, it is desirable to allow non-projective dependency structures. We show how a datadriven deterministic dependency parser, in itself restricted to projective structures, can be combined with graph transformation techniques to produce non-projective structures. Experiments using data from the Prague Dependency Treebank s...
متن کاملExtended Constituent-to-Dependency Conversion for English
We describe a new method to convert English constituent trees using the Penn Treebank annotation style into dependency trees. The new format was inspired by annotation practices used in other dependency treebanks with the intention to produce a better interface to further semantic processing than existing methods. In particular, we used a richer set of edge labels and introduced links to handle...
متن کاملTurning on the Turbo: Fast Third-Order Non-Projective Turbo Parsers
We present fast, accurate, direct nonprojective dependency parsers with thirdorder features. Our approach uses AD3, an accelerated dual decomposition algorithm which we extend to handle specialized head automata and sequential head bigram models. Experiments in fourteen languages yield parsing speeds competitive to projective parsers, with state-ofthe-art accuracies for the largest datasets (En...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2014