converting dependency structures to phrase structures for free-word -order languages

In this paper we propose an algorithm for converting dependency structures to phrase structures. The existing methods are mainly designed for languages with almost fixed word-order such as English. We first review these methods and on the basis of them we try to adopt a more general and flexible approach to the problem. We mainly concern the characteristics of free-word-order languages such as Persian. The algorithm is intended to bridge between an existing dependency-based representation for a free-word-order language and a well-studied constituency-based scheme. It can also be employed to automate the annotation process of a treebank in a way in which the user provides the dependency structure of a sentence and the algorithm automatically builds the desired phrase structure

کلیدواژه ها:

component ، Natural language processing ، treebanks ، depenency structure