Dependency Parser Of Chinese Long Sentence Based On Statistical Methods

Syntax analysis is one of the important issues of natural language processing. It aims to research the relationship of words and phrases and describe the roles words play in a sentence. For many years, the method of syntax analysis is based on the phrase structure. With the development of syntax analysis technology,the dependency syntax analysis shows its superiority gradually. This paper proposes a method to solve the problem of parsing Chinese long sentences based on the deterministic algorithm. First, according to the syntax analysis method of gradation, This paper divide sentences into two sub-sentences by the information of root. then, parse the two sub-sentences separately. By this way, the complexity of the sentences is declined in some way, thus, the parsing accuracy is increased obviously.Second, the parsing Strategy and method are improved in this paper. As to the parsing method, in order to solve the problem of Early-reduce caused by Nivre's Arc-eager algorithm, the two-phase parsing method proposed by Meixun Jin is further improved in this paper. The improved two phase parsing method focus not only on the verbs, but also focus on the Prepositions which easily cause the problem of Early-reduce in Chinese.At last, as to the paring direction, according to the character of projective of Chinese and the trait of sentences after segmentation, this paper proposes a method of using both the forward and backward directions. Experiments show that the parsing accuracy rate will be improved by using both the forward and backward parsing directions.