Dissertation > Excellent graduate degree dissertation topics show

Multi-Label Classification Based on Fuzzy Kernel Clustering and Fuzzy Support Vector Machine

Author: ZhengWenBo
Tutor: YangYan
School: Southwest Jiaotong University
Course: Signal and Information Processing
Keywords: Data Mining Pattern Recognition Multi - label classification Support Vector Machine Fuzzy Kernel Clustering
CLC: TP311.13
Type: Master's thesis
Year: 2011
Downloads: 88
Quote: 1
Read: Download Dissertation

Abstract


Since late last century, people entered the information age, a variety of data on the mass accumulation, far beyond the range of human processing \To this end, data mining technology came into being, and shows strong vitality. The classification is one of the most common tasks in data mining. It is summed with the information by a known law, is used to discriminate the new data, its prediction process. Special a sample with multiple label classification problem, referred to as single-instance multi-label classification problem. Unlike the common single-label classification data with multiple labels, in such matters, so that the samples attribution blurred, it is difficult to accurately classify, with considerable difficulty. However, its use is widespread in everyday life, many scholars are committed to this, there are many excellent algorithm and its improvement. In this paper, the problem of design of a multi-label classification algorithm based on fuzzy support vector machine. Support vector machines (Support Vector Machine, SVM) is a new sorting machine in the late 1990s by AT T Bell Laboratories Vapnik et al. The classification is based on statistical learning theory and structural risk minimization principle integrated optimal separating hyperplane kernel function, convex quadratic programming technology, which can be solved effectively \points and other issues, has good generalization and accuracy. Support vector machine is designed for two types of single-label data sets can not be directly applied to multi-class, multi-label. This paper designed a fuzzy support vector machine, able to contain two types of data, the sample may have two labels set of data classification. The classifier using fuzzy design ideas, through a membership function of the sample set, and make full use of the data information. This classification does not exist indecomposable area having a good classification precision. Designed based on distance and density in order to accurately describe the sample class affiliation, membership function. Take into account the special nature of the multi-label classification, this paper uses a one-to-one decomposition strategy, the raw data is decomposed into multiple two types of dual-label subset of training, and then use the voting method combination, the final completion of the multi-label classification. In order to improve training speed, to reduce the impact of noise in the training set points on the optimal decision hyperplane, the introduction of a fast fuzzy kernel clustering technology, to improve the performance of the algorithm. Finally, the results with some existing multi-label classification algorithm are compared to experiment in the experimental section, we first summarize the evaluation criteria are widely used multi-label classification algorithm, then the UCI datasets.

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 Text Classification Based on Biomimetic Pattern Recongnition,TP391.1
  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. Gao Zhong-ying academic thought and experience and use of Bufei Decoction treatment of common diseases of the respiratory system drug law,R249.2
  7. Bing- thick academic thought and clinical experience and empirical studies apply to turtle soups treatment of chronic kidney disease,R249.2
  8. The Design and Implementation of Bicluster Data Analyzing Software,TP311.52
  9. Data warehouse technology in the banking customer management systems research and implementation,TP315
  10. Based on Data Mining Technologies in Urban Water Supply Analysis and Decision,F299.24;F224
  11. Research on Face Recognition Based on AdaBoost Algorithm,TP391.41
  12. Liupanshui City tobacco company's human resource management system integrated design and implementation,TP311.52
  13. Research on Feature Extraction, Selection and Classification Algorithms for Pulmonary CAD,TP391.41
  14. Capsule endoscopy and endoscopic image portable receiver system bleeding Recognition Algorithm,TP391.41
  15. Data mining technology in the Expo traffic with social media Prediction Research and Application,TP311.13
  16. Based on Data Mining Technology Jobs System Design and Implementation,TP311.13
  17. Fault Diagnosis on Fuel Control System of a Certain Aero-engine,V263.6
  18. Decision Tree Based on Generalized Information Entropy Model and Its Performance Evaluation,TP18
  19. FSVM -based data mining method and its application to intrusion detection research,TP393.08
  20. CUDA-based video fire detection system,TP391.41
  21. Intrusion detection based on data mining technology research,TP393.08

CLC: > Industrial Technology > Automation technology,computer technology > Computing technology,computer technology > Computer software > Program design,software engineering > Programming > Database theory and systems
© 2012 www.DissertationTopic.Net  Mobile