Dissertation > Excellent graduate degree dissertation topics show

Improved Measures of Name Recognition in Chinese Text

Author: WuYuZhou
Tutor: SongRou
School: Beijing Language and Culture University
Course: Linguistics and Applied Linguistics
Keywords: Named Entity Recognition Name Recognition Reliability
CLC: H08
Type: Master's thesis
Year: 2007
Downloads: 192
Quote: 1
Read: Download Dissertation


Named entity recognition is a hotspot in recent research of Chineselanguage information processing. Named entity is the most important informationcarrier of texts. Increasing the accuracy of named entity recognition is helpful for bothincreasing the accuracy of word segmentation and improving the performance ofapplication systems, such as information extraction, information retrieval, machinetranslation and speech synthesis.Presently, there are three steps in the popular routine of named entity recognition:First, analyse the inner stucture and features of proper nouns, and get an informationtable. Second, train a proper language model based on a large list of proper nouns anda tagged corpus. Third, recognize the proper nouns in real texts using the languagemodel. However, due to the stability of inner structure and features of poper nouns,the existing recognition methods are becoming similar or even the same oninformation table of proper nouns. As far as name recognition is concerned, on theone side, most previous researches recognize names with statistical models based onsimple analysis of contexts, lacking of particular and deep mining. On the other side,there exist drawbacks in the probabilistic estimation of the popular methods based oncorpus or statistical language model, which can not reflect the language facts. In lightof these problems, we first improve the information table of name features, dig outsome helpful features, both positive and negative, and then calculate the reliability ofnames and their contexts using an improved method of reliability estimation.Experiments prove that the measures we took are helpful in the improvement ofname recognition. Our method can also provide some reference for the recognition ofother kinds of proper nouns.Wu Yuzhou (Linguistics and Applied Linguistics)Directed by Pro.Song Rou

Related Dissertations

  1. Development and Preliminary Application of the Questionnaire on Emotional Labour of the Middle School Teacher,B841
  2. The Research on Reliability of Low-Voltage Power Line Communication,TM73
  3. The Research and Application of Software Reliability Test for User Right Management System,TP311.53
  4. Research on Subsea Pipeline Repair Coupling,TE973
  5. Research and Application on Reliability Technology of Mine Hidrological Monitoring System,TD742.1
  6. Body Sensor Networks Adaptive Communication Protocol,TP212.9
  7. Periodic testing of the storage system reliability model,O213.2
  8. Amendment of Mental Health Evaluation Scale for Chinese Nuclear Power Plant Operators,B849
  9. Degradation of Amorphous Silicon Thin Film Transistors under Gate and Drain Voltage Bias Stress,TN321.5
  10. Research on Cross-document Coreference of Chinese Person Name,TP391.1
  11. Application of the Coal-powered Anti-grade Trip Protection System,TM732
  12. Improvement and Realization of AODV in Wireless Sensor Networks,TP212.9
  13. Web Page Attribute Extraction Method Research,TP391.1
  14. Reliability-based Sensitivity Research on Multi-span Rotor System,TH113
  15. Development of Reliability Analysis Software of the Tower Crane Boom,TH213.3
  16. High-Reliability Distributed iSCSI Storage System Based on Double-Stage RAID,TP333
  17. The Research and Implementation of Energy-efficient RAID,TP333
  18. Reliability and Security Study of the Digital Substation System,TM732
  19. DURO: A New Online Reconstruction Optimization for Single-disk-failure of RAID-6,TP333
  20. The Research for Named Entity Recognition and Relation Extraction in Text,TP391.1
  21. The construction of the thermal power plant enterprise information portal,F426.61

CLC: > Language, writing > Linguistics > Applied Linguistics
© 2012 www.DissertationTopic.Net  Mobile