Building Uyghur dependency treebank: Design principles, annotation schema and tools

Abstract

Treebank is a crucial source of information for NLP and linguistic researches. In this paper, we describe the process of building a Uyghur dependency treebank, including designing principles, annotation schemas and tools for corpus creation. The Uyghur Treebank is built from a public readings corpora, employed multi-tier representation for extending future use, and created about 23 dependency relations. This paper presents the preliminary results of this project and an overview of the new idea about combining this project with Language Grid.