Dissertation > Excellent graduate degree dissertation topics show

The Research on Svm-Based Method of Information Classification in Pervasive Computing Applications

Author: YangXiaoPeng
Tutor: GuoMinYi
School: Shanghai Jiaotong University
Course: Computer Software and Theory
Keywords: Pervasive Computing Support Vector Machine Sequential minimal optimization algorithm Semi - sparse algorithm Vector multiplication
CLC: TP181
Type: Master's thesis
Year: 2010
Downloads: 82
Quote: 0
Read: Download Dissertation

Abstract


Pervasive computing environment, contextual information analysis is very important to its ability to provide the right service Pervasive Computing environment plays a decisive role. Pervasive computing applications require fast and accurate contextual information classification and management, however, often a huge amount of information collected, the variety has a very important significance to find an effective classification used in pervasive computing environment . The support vector machine classification is a machine learning method based on statistical learning theory, nonlinear and high dimensional samples training showed a unique advantage. Support vector machine classification method to its theoretical advantages of support vector machine classification method has achieved outstanding results in the field of text classification applications, at the same time there are a wide range of research and applications in other areas, such as face recognition and image processing. A variety of support vector machine algorithm for vector multiplication semi-sparse algorithm, and used in sequential minimal optimization methods to improve the speed of operation of large-scale sparse matrix vector multiplication, in order to optimize SVMTorch computational performance of the classifier. By theoretical analysis, two each containing the vector of m and n elements for comparison and addressing, the use of traditional sparse algorithm SVMTorch algorithm requires O (mn) of the time consumption, while the semi-sparse algorithm can be O ( n) time to complete the multiplication processing of these two vectors, and does not affect the accuracy of support vector machine classifier. The experimental results show that the performance is significantly better than original SVMTorch classifier performance based on the semi-sparse algorithm SVMTorch classifier. WebKB and 20-newsgroup corpus, based on half sparse the algorithm SVMTorch training time were 54.32% and 74.95% of the original SVMTorch. In addition, support vector machine SVMTorch classifier to expand, making it not only support of multiple classifiers single label classification problems, and also supports the classification of multi-label classification problems by update SVMTorch classifier training and testing inspection algorithm, its output to support multi-label classification computation and multi-classification label function on the Reuters-21578 corpus verified. In order to further improve the performance of the calculation of the SVMTorch classifier, the article uses the messaging interface model (MPI) parallelization of the SVMTorch classifier, so that it can be calculated in parallel on multi-core processors and a distributed cluster. Eventually achieve a semi-sparse algorithm-based multi-label classification, the parallelization SVMTorch function prototype and applied to Chinese web page classification.

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. The Research on Intrusion Detection System Based on Machine Learning,TP393.08
  10. Research on Improved K Neighbor Support Vector Machine Algorithm Faced Text Classification,TP391.1
  11. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  12. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  13. Philosophical Research on the Intelligence Space in the Pervasive Computing Environment,N02
  14. Research on Subimage Selection and Mathching Method for Synthetic Aperture Radar(SAR) Target Recognition,TN957.52
  15. Research of Facial Expression Recognition Algorithm,TP391.41
  16. Fundus Image Segmentation Based on SVM and Template Matching,TP391.41
  17. Research and Realization of License Plate Character Recognition Algorithm Based on SVM,TP391.41
  18. Cloud Classification Based on Geostationary Meteorological Satellite Imagery,TP391.41
  19. The Studies on Some Improvements of the GA and Their Applications in SVM,TP18
  20. The Research and Implementation of the Fault Management System for Triple Play,TP315
  21. Research on Context Reasoning and Inconsistency Detection in Pervasive Computing,TP391.1

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