Dissertation > Excellent graduate degree dissertation topics show

A K-MEANS Clustering Algorithm Based on ALPHA-STABLE Distribution

Author: XuMingZhe
Tutor: ZhangJunYing
School: Xi'an University of Electronic Science and Technology
Course: Applied Computer Technology
Keywords: K_MEANS Clustering algorithm ALPHA-STABLE distribution Gaussian distribution Fractional lower order moments p - norm
CLC: TP391.41
Type: Master's thesis
Year: 2008
Downloads: 56
Quote: 0
Read: Download Dissertation

Abstract


ALPHA-STABLE distribution of a large number exist in real life , it is currently one of the international study comparing popular topics ALPHA-STABLE distribution a good description of the distribution of real-world data , it has gradually been applied to a variety of fields , nature more and more people understand . This article is mainly for obedience ALPHA-STABLE distribution data clustering analysis . In recent years , the study of the clustering algorithm has achieved considerable development , including K-MEANS algorithm has been widely used in its algorithm is simple , efficient performance . This paper studies found that K-MEANS algorithm suitable for clustering obey variance Gaussian distribution of data , statistical significance , it is mainly the use of second-order statistics of the data . But to obey the ALPHA-STABLE distribution data , the ALPHA bands ( ALPHA lt; 2 ) as well as higher-order statistics is infinite , the paper argues that the K-MEANS algorithm is not suitable to do clustering obey ALPHA-STABLE distribution data directly analysis . Based on the above analysis , this paper presents a K-MEANS clustering algorithm based ALPHA-STABLE distribution . The algorithm is based on the following ideas : thought K-MEANS clustering algorithm based on fractional lower order moments ideological similarity measure that p - norm ; measure p - norm Gaussian data distribution , which can take advantage of the idea of K-MEANS algorithm indirect clustering data , this would resolve the K-MEANS clustering algorithm for problem obey ALPHA-STABLE distribution data clustering . Experimental results show that it is applied to the simulation data and real data , clustering algorithm proposed in this paper to improve the clustering performance .

Related Dissertations

  1. Stability Analysis and Controller Design for Discrete-Time Switched System,TP13
  2. Integrated Social Contract,B82-05
  3. Citizen Participation of Urban Inhabitant: Structure and Measurement,B849
  4. Research on K-means Optimization Clustering Algorithm,TP311.13
  5. Research on Fuzzy C-Mean Clustering Algorithm Based on Particle Swarm Optimization and Shuffled Frog Leaping Algorithm,TP18
  6. Research on Clustering Algorithm Based on Genetic Algorithm and Rough Set Theory,TP18
  7. The Research on Routing Protocol of Agricultural Environmental Monitoring System Based on Wir Eless Sensor Networks,TN915.04
  8. A Psychological Experiment Research on Using Descriptive Norm to Arose the Public Behavior,X2
  9. Multiple Intelligence Assessment Scale pupils norm established National City,G449.5
  10. SAR interferometric method for optimal selection,P225.2
  11. The Criterion of the Fully K-convexity and Locally Uniform Rotundity of Orlicz-Lorentz Spaces with the Orlicz Norm,O177
  12. Segmentation of cDNA Microarray Image Using Fuzzy C-means Algorithm Optimized by Particle Swarm,TP391.41
  13. The Research of Kernel Adaptive Filtering Algorithms,TN713
  14. Research on Routing Algorithmin Sensor Networks Based on Cluster with Mobile Sink,TP212.9
  15. Research on Spectrum Detection Technology from Compressed Sensing,TN925
  16. An Investigation Study on Quality of Life and Related Factors in Hubei Province,F126;F224
  17. Psychological problems of patients with gynecological malignancies and its norm Assessment Scale Development,R473.73
  18. World Bank and Chinese low - carbon economic development,F124;F205
  19. Static and Dynamic Stability Analysis of Slope Based on Three-Dimension Nonlinear Finite Element Method,TU43
  20. Coordinate Systems in the Informationized Surveyingand Mapping and Itsapplication Research,P228.4
  21. Some New Kinds of Solutions to the Iterative Boolean-like Functional Equation,O159

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer applications > Information processing (information processing) > Pattern Recognition and devices > Image recognition device
© 2012 www.DissertationTopic.Net  Mobile