Multiword Expression Translation Using Generative Dependency Grammar
نویسنده
چکیده
The Multi-word Expressions (MWE) treatment is a very difficult problem for the Natural Language Processing in general and for Machine Translation in particular. This is true because each word of a MWE can have a specific meaning but the expression can have a totally different meaning both in source and in target language of a translation. The things are complicated also by the fact that the source expression can appear in the source text under a very different form from its form in a bilingual MWE dictionary (it can have some inflections) and, most of all, it can have some extensions (some MWE words can have associated new words that do not belong to the MWE). The paper show how this kind of problems can be treated and solved using Generative Dependency Grammar with Features.
منابع مشابه
GRAALAN – Grammar Abstract Language Basics
This paper gives an outline about most important features of GRAALAN (Grammar Abstract Language) used for linguistic knowledge description. GRAALAN is an implementation of theoretical concepts of GDGF (Generative Dependency Grammars with Features) and AVT (Attribute Value Trees). GDGF is based on dependency trees (DT) and a generative process. GDG eliminates some issues of Dependency Grammars D...
متن کاملA Generative Dependency Grammar
This document presents a new kind of grammar: the Generative Dependency Grammar (GDG). This type of grammar is based on dependency trees (DT) and a generative process. GDG will eliminate some issues of DG (by example the missing of phrasal categories) and GG (the problem of discontinuous structures) and will merge the advantages of the two types of grammar (GG the representation of phrasal cate...
متن کاملMultiword Expressions As Dependency Subgraphs
We propose to model multiword expressions as dependency subgraphs, and realize this idea in the grammar formalism of Extensible Dependency Grammar (XDG). We extend XDG to lexicalize dependency subgraphs, and show how to compile them into simple lexical entries, amenable to parsing and generation with the existing XDG constraint solver.
متن کاملSemi-Automated Resolution of Inconsistency for a Harmonized Multiword Expression and Dependency Parse Annotation
This paper presents a methodology for identifying and resolving various kinds of inconsistency in the context of merging dependency and multiword expression (MWE) annotations, to generate a dependency treebank with comprehensive MWE annotations. Candidates for correction are identified using a variety of heuristics, including an entirely novel one which identifies violations of MWE constituency...
متن کاملSynchronous Dependency Insertion Grammars: A Grammar Formalism For Syntax Based Statistical MT
This paper introduces a grammar formalism specifically designed for syntax-based statistical machine translation. The synchronous grammar formalism we propose in this paper takes into consideration the pervasive structure divergence between languages, which many other synchronous grammars are unable to model. A Dependency Insertion Grammars (DIG) is a generative grammar formalism that captures ...
متن کامل