Dissertation > Excellent graduate degree dissertation topics show

Composite Parameters Selection for Support Vector Machines

Author: JiaLei
Tutor: LiaoShiZhong
School: Tianjin University
Course: Applied Computer Technology
Keywords: Support Vector Machine Model selection Radius / interval boundary Parameter adjustment
CLC: TP18
Type: Master's thesis
Year: 2007
Downloads: 231
Quote: 0
Read: Download Dissertation

Abstract


Parameter selection of support vector machine directly determines the efficiency and the effect of training support vector machines , support vector machine model selection of the important issues . Existing support vector machine parameter selection method using the nested two - loop optimization process , the calculation process , first training the classifier , and then update the parameters . Number of iterations such methods exist , the problem of poor selection effect . Parameter selection of support vector machine this problem , and to achieve a combination of parameters selected specific work includes : 1 . Adjust parameters and classifier combinatorial optimization framework . The framework of the application of heuristic search methods search parameter space , and in the same iterative process to adjust the classifier combinatorial optimization of parameters and classifier . 2 . Derive a new boundary radius / interval . The sector is an approximation of the radius / interval profession statistical learning theory , under the premise of ensuring the accuracy of learning to improve computational efficiency . 3 are given the choice of the initial parameters . From the theoretical analysis of the asymptotic properties of the kernel function parameters in the two extreme cases , parameter selection principle , given the parameters of the initial point selection method . 4 . Combination of parameter adjustment algorithm is designed and implemented . Radius / Interval sector , new derivation of the algorithm is applied to a combination of sequential unconstrained minimization and variable metric method two optimization algorithms , given the parameters of the optimal solution . 5 . Carry out experimental research . In the UCI database Heart , Diabetes , A2a, W1a of and Statlog database German.Numer five standard datasets , test and compare the regulation of cross-validation of the method and the traditional combination of parameters , the gradient descent method performance . The experimental results demonstrated the effectiveness of a combination of methods .

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 for Infrared Image Target Identification and Tracking Technology,TP391.41
  7. Study on the Road Condition Monitoring Based on Vehicular 3D Acceleration Sensor,TP274
  8. Research of Diagnosing Cucumber Diseases Based on Hyperspectral Imaging,S436.421
  9. Bayesian methods under two ordinal value of multi-value data model with simultaneous identification of outliers,O212.1
  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. Random Network Model Discrimination,O157.5
  14. Parameter Optimization Research in GSM and TD-SCDMA System Switching,TN929.53
  15. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  16. Research on Subimage Selection and Mathching Method for Synthetic Aperture Radar(SAR) Target Recognition,TN957.52
  17. Research of Facial Expression Recognition Algorithm,TP391.41
  18. Fundus Image Segmentation Based on SVM and Template Matching,TP391.41
  19. Research and Realization of License Plate Character Recognition Algorithm Based on SVM,TP391.41
  20. Cloud Classification Based on Geostationary Meteorological Satellite Imagery,TP391.41
  21. The Studies on Some Improvements of the GA and Their Applications in SVM,TP18

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