Dissertation > Excellent graduate degree dissertation topics show

Corpus - based clutch Words

Author: WangChunXia
Tutor: ChenXiaoHe
School: Beijing Language and Culture University
Course: Linguistics and Applied Linguistics
Keywords: Separable Words Moving morpheme Name morpheme Combination Model Rule Probability
CLC: H146.2
Type: Master's thesis
Year: 2001
Downloads: 822
Quote: 13
Read: Download Dissertation

Abstract


This article through the investigation and analysis of large-scale corpus has been clutch word appear in the text and insert the component of the law as a form of a special language of these laws are summarized clutch word with the group Clutch word mode, there is no group with mode manually write their insertion rules. On this basis, a combination of rules and statistics algorithm, marked the acrostic is a closed test and open test, open test results: the correct rate of 81.74%, 98.27% recall rate. The text is divided into six parts: Part I: Introduction. Define some of the concepts of acrostic to determine the goals and methods of the topics, pointed out that the value and significance of the research, and Clutch word processing status and the status of academic research in the field of linguistics and natural language information, and derive some inspiration. Corpus used in this study. Part II: the difficulty of the acrostic is marked. The the sentence corpus acrostic has a general understanding, preliminary processing of the corpus acrostic is, a statistical analysis. Pointed out the difficulties in the clutch Words and advantages according to the sentence. Part III: Clutch word insertion rules to obtain and analyze. This part is to determine the basis of the algorithm. We summarize the group with mode according to the mode effective rules as well as other data obtained from a large number of clutch word sentences, in order to make up for data sparse insufficient, artificial summarizes part of the acrostic rules. Part IV: data and rule-based algorithms and labeling. This part is the core of the research work, use the previously obtained data and the rules of the marked acrostic algorithm, the algorithm is described. Closed test and open test, respectively, using the training corpus and test corpus, and has made some adjustments in the labeling process of the algorithm. Part V: marked results reporting and analysis. Shows the the label closed test and an open test results and marked instance, and marked the results of the comprehensive analysis and comparative analysis, analysis of typical errors in the labeling, summed up the difficulties of the present study. Part VI: Conclusion. Comprehensive summary of the research, some shortcomings in the research, put forward the need to improve in the next step. The study is valuable on certain data values ??and algorithms in natural language information processing, experimental results also raise the possibility of the future can continue to study.

Related Dissertations

  1. Research on Orbital Control Method for Space Rendezvous and Docking,V526
  2. Research on Effectiveness Evaluation Methods of Weapon Systems,TJ06
  3. Research on Dam-Break Probability Analysis Method for Dangerous Reservoir,TV697
  4. Key Technologies of Dam-break Life Loss Risk Assessment Research,TV122.4
  5. Research on Navigation System Related Technology for Moving Objects under Dynamic Environment,TP301.6
  6. Improvement of Ant Colony Algorithmand Its Application in Robot Path Planning,TP242
  7. Xu Peng age of academic thought and clinical experience medication regularity of the treatment of chronic gastritis,R249.2
  8. Research on Athletics Gymnastics Horizontal Bar Technology Development Regularity,G832
  9. On Ecological Politics of Murray Bookchin,D09
  10. Zhou Dean of academic thought and clinical experience and treatment of Tourette Syndrome the acupoints drug laws,R246
  11. Research on Legalization of Political Power Operation from the Vision of the Rule of Law,D920.0
  12. The Comparison and the Consideration about the Difference in Curriculum Standards Content in Statistics and Probability between China & England,G633.6
  13. Current Status of Nutrient Input and Soil Nutrient Content Change Characteristics in Solar-greenhouse in Quwo County,S626
  14. Evolution of Volleyball Competition Rules modify the impact of the development of volleyball,G842
  15. Wide dependency structure and tail probability random Asymptotics,O211.5
  16. Linearly polarized pulsed laser fields lithium atom Rydberg transitions population,O562
  17. Studies on the Public Service Barrister System of China,D926.5
  18. A Design and Implementation of Single Sign-on System Based on the Improved RBAC Model and CAS,TP311.52
  19. The Research of Bank Accounting Query System Supporting Top-k Query,TP311.52
  20. Study on Skyline Query Processing Techniques on Uncertain Data,TP311.13
  21. On the legal regulation of private lending,F832.4

CLC: > Language, writing > Chinese > Grammar > Modern,modern syntax > Part of speech
© 2012 www.DissertationTopic.Net  Mobile