Building contemporary uyghur grammatical information dictionary

Abstract

“Contemporary Uyghur Grammatical Information Dictionary” is the basic language knowledge base for the Uyghur information processing. It provides a large amount of grammatical information and collocation features for 49,072 words. The original intention of the development of Uyghur grammatical information dictionary is to provide basic resources for Natural Language Processing (NLP). Building information dictionary has far-reaching theoretical and practical value for Uyghur text retrieval, proofreading, machine translation, summary generation, linguistic knowledge acquisition, representation and usage, even allow the computer to “understand” language. In this paper, we use the methods of computational linguistics, corpus linguistics and NLP techniques to analyze Uyghur morphology, Uyghur syntax. On this basis, we study the grammatical features of Uyghur nouns, verbs, and adjectives and so on, and then establish classification system of part of speech of Uyghur. Guidance with this classification system, we use relational database technology to design structures of “Contemporary Uyghur Grammatical Information Dictionary”. According to the principle of combining grammatical functions and meanings, using methods of corpus linguistics, we select words from contemporary balanced Uyghur corpus, and import them to the “Uyghur Grammatical Information Dictionary”, then give each word’s grammatical attributes. Finally we build “Uyghur Grammatical Information Dictionary” of practical value.