Dissertation > Excellent graduate degree dissertation topics show

SVM-based news features new event detection research

Author: ZhouShuXian
Tutor: LiuZhenYu
School: Central South University of Forestry Science and Technology
Course: Applied Computer Technology
Keywords: New Event Detection Support Vector Machine Vector space model
CLC: TP18
Type: Master's thesis
Year: 2011
Downloads: 37
Quote: 0
Read: Download Dissertation

Abstract


With the popularity of Internet news site has become a major source of access to information is one. But faced with the vast amount of information that people need help in some way to quickly and accurately collect information of interest. Topic Detection and Tracking (Topic Detection and Tracking, TDT) is a massive response to the problem of information studies undertaken. United States in 1996 began to organize the relevant research institutions to conduct TDT evaluation, the new event detection (New Event Detection, NED) as one of its key sub-tasks, the main content is how the new event detection system via the first news reports of events . Although domestic research started late, but the topic detection and tracking in information retrieval, text classification, and other related fields of practical value, the related research has become a hot field of information processing. The basic system consists of a new event detection text representation, text, text classification and comparison with the time window to select several modules. In this paper, based on the research at home and abroad, the Chinese news in text form data stream is processed on natural language processing techniques unified expression time information, the standardization process names, names the effective establishment of the library has been improved so square. In particular, considering the news coverage tends to be defined as soon as possible to inform the reader, this paper presents the concept of location information, and on this basis can be added through the functions and multiplicative function on the Okapi similarity comparison function improvements. Text classification, the paper used in the SVM (Support Vector Machine, SVM) method has a solid mathematical foundation, the field of machine learning is a very high degree of concern one way, especially in unsupervised situations with good robustness, has been successfully applied to pattern recognition and regression analysis of the problem. In this paper, from WASHINGTON, Xinhua and other large sites collect contains 20 topics 14295 News Documents were used kernel regression method (kernel regression), nearest neighbor method (nearest neighbor method) and support vector machine method (Linear-kernel SVM, RBF-kernel SVM) methods such as new event detection, by using the basic NED NED systems and improved detection system experimentally derived coefficients consideration compared three methods to get the price corresponding coefficients were reduced by 4.9 per cent , 7.3% and 13.1%. NED system which describes the application of corpus-based news feature vector space model and the improved use of location information reflects the vocabulary Okapi similarity comparison function, the use of SVM method for text classification compared to kernel regression method and the nearest neighbor method classification, NED system performance more effective.

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. Research on Autamatic Music Structrue Analysis,TN912.3
  3. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  4. Fault Diagnosis Method Based on Support Vector Machine,TP18
  5. Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
  6. Research on Focused Crawler Based on SVM Classification Algorithm,TP391.3
  7. Research for Infrared Image Target Identification and Tracking Technology,TP391.41
  8. Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
  9. Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
  10. The Research on Intrusion Detection System Based on Machine Learning,TP393.08
  11. Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
  12. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  13. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  14. Research on Subimage Selection and Mathching Method for Synthetic Aperture Radar(SAR) Target Recognition,TN957.52
  15. Design and Implementation of the Character Classification System Used in Search Engine,TP391.3
  16. Research and Application on Short Message Text Clustering,TP391.1
  17. Research and Realization of License Plate Character Recognition Algorithm Based on SVM,TP391.41
  18. The Studies on Some Improvements of the GA and Their Applications in SVM,TP18
  19. The Research and Implementation of the Fault Management System for Triple Play,TP315
  20. Research for Event Extraction Method in Specific Domain Based on Tree Conditional Random Field,TP391.1
  21. Modulation Classification Algorithms of Digital Communication Signals,TN914.3

CLC: > Industrial Technology > Automation technology,computer technology > Automated basic theory > Artificial intelligence theory
© 2012 www.DissertationTopic.Net  Mobile