本文从自然语言处理的角度探讨俄语中NP构句块的模式化和自动识别问题，内容涉及NP构句块模式化的语言学基础、NP构句块的模式化及模式的多层级、多平面形式化描写(包括带有扩展模式形容词的模式化)、NP构句块自动化处理，包括NP构句块处理的理论前提、模式组配规则、NP构句块的边界测定、带后置定语的NP构句块识别等。并最终通过实例论证基于规则自动识别NP构句块的可操作性。 俄语NP构句块的自动化处理起着承接俄语自动形态略论和句法-语义略论的影响，是俄语自然语言处理的重要处理模块。本文的探讨在理论上，俄语论文范文，可揭示名词性短语的结构及功能属性，系统把握形容词扩展成分的复杂性和多构性，俄语论文题目，进一步揭示语义搭配的规律性；在操作层面上，通过构建俄语NP构句块模式和设计识别算法，可以最终实现俄语NP构句块的自动化识别，为俄语句子的自动句法-语义略论提供语言学保障，有助于提高句法略论器的略论质量。
This paper disertation explores on modeling and automatic recognition of NP in Russian sentences in light of Natural Language Processing(NLP), including linguistic foundation in the course of modeling, modeling of NP, multilevel description of models, modeling of AP with patulous components and automatic recognition of NP. In the process of recognition attention is paid to theoretic precondition of automatic recognition of NP, combinational rules of models, boundary mensuration of NP, automatic recognition of NP with postpositive attributes. Final paper of the disertation mechanism of automatic recognition of NP based on models is demonstrated with a concrete example.Automatic recognition of NP is regarded as an important step of NLP, connecting the automatic morphological analysis with the syntax-semantic analysis. In theory, this paper can show structural and functional attribute of NP, grasp complexity and multiformity of AP with patulous components, and then highlight rules of semantic combination. In practice, by constructing models of NP in Russian sentences and designing recognizable algorithm we can ultimately realize automatic recognition of NP, provide the syntax-semantic analysis of Russian sentences with linguistic guarantee and improve analytic quality of syntactic analyzer. |
