The Automatic Identification Research Of Preposition "Dao" And Structure For Information Processing

The preposition"dao"evolved from the verb, Language scholars has been widespread concern the part of speech and correlation structure processing,However,they have different viewpoints.Have created puzzled for the language teaching and the research,also does not favor the information processing.This aeticle take 1990 to 2002"Writer Digest Newspaper"as the research language materials,and take the chinese lexical anslysis system of Academia Sinica Computer Institute as the tool.We study the issue of the part of speech,then recognize it,finally,recognize the border of prepostitional structure. This article is divided into five chapters:Chapter1 is the introduction part.Introduce the significance of the topic and predecessor's research results, and propose own research technique learn from the predecessor's research, goal and Corpus processing.Chapter2 is about the"dao"distinguish between verb and preposition. Take the real language materials as the foundation, according to the reference booksa,special pre-component and post-component,and ingredients to distinguish, provide knowledge resources.Chapter3 discusses the part of speech segmentation of"dao".We get the conditions of"dao"when it is morpheme,verb or preposion,then make participle and labelling through vocabulary and disambiguation rule,we obtain the rule base of recongintion.Chapter4 is includes the boundary identification of"dao+X".According to the boundary characteristic of the preposition structure and the mutual information theory, we summarize the rule and carry on the boundary recognition.Chapter5 is the epilogue.it contains the conclusion and future study.