Dissertation > Excellent graduate degree dissertation topics show

Text-oriented classification method of feature word selection

Author: ChenJian
Tutor: HuXueGang；WangWeiWei
School: Hefei University of Technology
Course: Computer technology
Keywords: Text Classification Feature Selection Feature Extraction Support Vector Machine Classifier
CLC: TP181
Type: Master's thesis
Year: 2009
Downloads: 132
Quote: 1
Read: Download Dissertation

Abstract


With the development of information technology , a large number of text data generation and led to the rapid spread of the requirements for text classification . As one of the key technologies for text classification feature selection , the classic feature selection method does not take into account the distribution of entries in the category , mostly based on the threshold simply remove low-frequency words , and information retrieval theory that \document feature large contribution . \To solve these problems , this paper, on the classical TF-IDF feature selection method based on the analysis method is proposed term evaluation balanced distribution feature word selection method , on the basis of a feature vector is constructed as a weighted weighted number classifier , the final SVM based text classification algorithm validate its effectiveness . In this paper are as follows: ( 1 ) an overview of text classification . Analysis of the text classification and support vector machines for text classification and the basic theory of development , the main line of text categorization , text classification techniques were analyzed research on key technologies of the classification process - text feature selection methods and classification algorithms in detail comparative analysis ; ( 2 ) the word balanced distribution proposed evaluation feature word selection. In the classical method of feature selection algorithm DF detailed analysis of the problems , based on the balance of the word proposed evaluation feature word selection. By considering the entries within the class probability and probability of occurrence within the class documentation , evaluation of the entries in the class distribution within the text of the equalizer as the main basis for feature word selection . Balanced distribution of feature words the smaller , indicating that within the class and within-class document distribution is more balanced, more able to represent such characteristic words . Experiments show that the method in classification accuracy has been greatly improved ; ( 3 ) Weighted classifier structure. The number of samples for the experiment is not balanced in the case , resulting in a balanced experimental classification effect is not the problem , on the classifier construction method were analyzed , proposed a weighted classifier, namely the statistics of each class the number of training vectors set of samples to the number of feature vectors as weights and use text classification based on support vector machine classifier constructor method comparison experiment , the experiment proved the effect of weighting classifiers tends to be more balanced .

Related Dissertations

  1. Research on Automatic Detection Algorithm for Substructure Distress of Highway Pavement Based on SVM,U418.6
  2. ISAR Imaging Simulation of Space Targets and Target Recognition Based on ISAR Images,TN957.52
  3. Research on Autamatic Music Structrue Analysis,TN912.3
  4. Research on Feature Extraction and Classification of Pulse Waveform for Cholecystitis and Nephrotic Syndrome Diagnosis,TP391.41
  5. Application of Q-Learning in the Content-Based Image Retrieval Technology,TP391.41
  6. Research on Transductive Support Vector Machine and Its Application in Image Retrieval,TP391.41
  7. Research on Feature Extraction and Classification of Tongue Shape and Tooth-Marked Tongue in TCM Tongue Diagnosis,TP391.41
  8. Research on Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  9. Tourism Comments on the Internet’s Semantic Analysis and Usefulness Research,TP391.1
  10. Research on Visual Measurement for Spacecraft Rendezvous and Approach,TP391.41
  11. Research on the Image Real-Time Acquisition, Storage and Image Processing System,TP391.41
  12. Feature Extraction, Selection and Combination in Lipreading,TP391.41
  13. Multi-currency Notes Technology Research and Implementation,TP391.41
  14. The Research on Paper Currency Classification Method Based on Harr-Like Feature and Minimal Ball Including Samples,TP391.41
  15. Pavement Distress Recognition Based on Image,TP391.41
  16. Research on Visual Detection and Tracking of Mobile Robots,TP242.62
  17. Research on Fusion Algorithm of Hyper Spectral and High Spatial Resolution Remote Sensing Image,TP751
  18. Fault Diagnosis Method Based on Support Vector Machine,TP18
  19. Process Support Vector Machine and Its Application to Satellite Thermal Equilibrium Temperature Prediction,TP183
  20. An Approach for Identifying a Plant Resistance Gene Based on the Random Forest,Q943
  21. Tobacco Diseases Auto-Recognition Research Based on Image Processing Technology,S435.72

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