Dissertation > Excellent graduate degree dissertation topics show

Two Semi-supervised Dimensionality Reduction Methods for Tensor and Multiple View Data

Author: HuKui
Tutor: WuZuo
School: National University of Defense Science and Technology
Course: Applied Mathematics
Keywords: Tensor data Multiple views of data Harmonic functions Semi-supervised Dimensionality reduction
CLC: TP181
Type: Master's thesis
Year: 2010
Downloads: 27
Quote: 0
Read: Download Dissertation

Abstract


High-dimensional data dimensionality reduction statistics and computer science as a crossover study direction , has become a hot research . Semi-supervised dimensionality reduction because it can take advantage of unlabeled data, the same data can be obtained on the general supervision approach is more desirable than the dimensionality reduction effect , in recent years has been more and more attention of researchers . However , the practical application of high-dimensional data in addition to outside , often also has a natural internal structure , the traditional semi-supervised dimensionality reduction methods are often faced with such data ineffective . Tensor structure and multi-view data structures are two typical structures . In this paper, a label based communication technologies harmonic function on the basis of the data presented for the tensor data and multi-view semi-supervised dimensionality reduction method . In this paper, complete the following two tasks: 1. Nie et al in local tensor discriminant analysis (LTDA) , based on the proposed harmonic functions based semi-supervised tensor dimensionality reduction method (TDRHF), the first use of this method to reconcile function to a small label with the tag data is not transmitted to the tag data , and then the data is semi-supervised dimensionality reduction, semi-supervised algorithm of the Nie et al 's work extended to semi-supervised areas . 2 In Hou et al proposed multi- view data semi-supervised dimensionality reduction method (MVSSDR) , based on the proposed multi- harmonic functions based view data dimensionality reduction method (MVDRHF), this method will MVSSDR data used the connection between the label information to propagate through the harmonic function after soft label information , avoiding the tag information and connection information between the transformation required to keep a priori information , to obtain a more desirable effect of dimensionality reduction .

Related Dissertations

  1. Improving of Artificial Imune Classification and Anomaly Detection Algorithms,R392.1
  2. Research on Image Super-resolution Reconstruction Based on Non-local Similarity,TP391.41
  3. Semi-supervised Learning and Active Learning of Sentiment Classification Coupled with Domain Knowledge,TP181
  4. P2P Traffic Identification Method,TP393.06
  5. Semi-supervised hashing algorithm based Image Retrieval Methods,TP391.41
  6. TS201 -based space-time adaptive processing algorithms and engineering to achieve,TN957.51
  7. The LVCSR system based on adaptive methods of semi-supervised learning,TN912.34
  8. Research of Anomaly Detection Algorithms of Hyperspectral Imagery Based on Source Data Optimized,TP751.1
  9. Research on Intrusion Detection Based on Semi-Supervised Fuzzy Clustering,TP393.08
  10. Study on Clustering and Classification with Applications,TP18
  11. Dimensionality reduction for nonlinear systems Strategy and Simulation,TP11
  12. Dimensionality reduction algorithm based on subspace biometric Applied Research,TP391.41
  13. Semi-supervised discriminant analysis method,TP391.41
  14. Research on Face Feature Extractions Based on Locality Preserving Projections,TP391.41
  15. Based on independent component analysis for image retrieval research,TP391.41
  16. Graph-based semi-supervised learning algorithm,TP181
  17. Based on semi-supervised integration of remote sensing image segmentation and classification,TP751
  18. Research on Laplacian Weighted Clustering Algorithms,TP18
  19. Oversight and semi-supervised canonical correlation analysis and its application,TP181
  20. Measure the distance learning theory and applied research,TP181

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