Conditional Random Fields Based Location Name Recognition In Ancient Chinese

Named entity recognition is the one key task in natural language processing and computional linguistics. Its research and application have great theoretical and practical significance. However, the current research in this area are mostly concentrated in the area of Modern Chinese. Named entity recognition in the Ancient Chinese is the key to improve the course of ancient books information processing, it desrve to be studied. This paper will make the use of a statistical model as Conditional Random Field (CRF) to do the research on the location name in the Ancient Books . In detail, this dissertation has conducted the following researshes:Chapter 1: IntroductionThis chapter explained the goal and significance of this research. Revolved the central task, introduce the related basic theory and the research background, were mainly in the natural language understanding meaning and thought. also has summarized the research present situation and the prospect about the ancient electronic books information processing.Chapter II: Named entity recognition and its methodFirst guide the question of named entity recognition, then introduced the significance and the method of this reseach, mainly based on the statistical method.Chapter III: CRF modelthe CRF(Conditional Random Fields) model is the main method in this article that studies the location name automatic recognition in ancient books. This chapter introduction the CRF model theory, lay emphasis on this model to apply in the named entity recognition.Chapter IV: CRF model based location name automatic recognition——take the "Romance of the Three Kingdoms" as an example this chapter is the hard core of this paper. take "the Romance of the Three Kingdoms" as the example, divided 13 steps three stage to introduce the experimental study process that using the CRF model to automatic identify the location name in the ancient books in detail.Chapter V: Conclusion:Carries on the summary finally to this research work, and make the forecast to present's following research.