Study On The Distribution And Statistical Of The Structure Tibetan Noun Phrases

Large data strategy and in-depth learning methods have become the mainstream of Tibetan natural language processing technology.At present,the lack of knowledge resources and annotated corpus have affected the process of intelligent research in Tibetan language,especially like the resources such as WordNet,HowNet,frame semantics,lexical semantic,syntactic structure annotation,semantic role labeling and text information labeling.Not formed a unified normative model,in-depth learning and other mainstream learning methods can’t be used for practical training.Therefore,the construction of resource library has become a basic and arduous task in the field of Tibetan information processing.It is the core problem to study noun phrases,verb phrases and adjective phrases in the construction of syntactic trees.In this paper,the classification of Tibetan noun phrase and its structure is carried out under the framework of Tibetan syntactic tree.The purpose is to test the accuracy of classification of Tibetan phrase structure and improve the efficiency of Tibetan phrase analysis,Library construction process.The article is divided into eight chapters to narrate.Firstly,it discusses the research background and research status of the phrase,and further studies the relevant syntactic analysis theory of noun phrases in English and Chinese and the corpus needed to construct the noun phrase structure library.Secondly,the concept of noun phrase phrases in English.Chinese and Tibetan is described,and the structure of noun phrases in Tibetan language is analyzed by the authentic corpus of Tibetan language.The noun phrases composed of classifiers are classified and classified by classification established a set of Tibetan phrase phrases.Finally,through the statistical results of the noun phrase structure in the Tibetan real corpus,we construct the structure of the noun phrase structure,the noun phrase label and the noun phrase annotation software.The article mainly adopts the corpus empirical,comparative analysis,statistical analysis,artificial annotation and artificial proofing research method.The basic noun phrase structure of the Tibetan language and the attributive annotation corpus are established.In summary,the classification and statistical research of Tibetan phrase structure types provide the basic resources for semantic analysis and tree building construction of Tibetan language,which provides some theoretical and technical support for information retrieval,search engine,machine translation,text classification,pattern recognition,multimedia teaching,network and other application technology fields.