A Winch Fault Classification Algorithm Based on Cluster Kernel Semi-Supervised Support Vector Machine
| Autor: | Huan Zhao, Xiao Xiao Kong, Chang Jian Zhu, Zhong Xiang Zhao, Xian Xin Shi, Li Jing Li, Jun Fei Chai |
|---|---|
| Rok vydání: | 2012 |
| Předmět: |
business.industry
Pattern recognition General Medicine ComputingMethodologies_PATTERNRECOGNITION Kernel method Polynomial kernel Kernel embedding of distributions Variable kernel density estimation Kernel (statistics) Radial basis function kernel Mean-shift Artificial intelligence Tree kernel business Algorithm Mathematics |
| Zdroj: | Applied Mechanics and Materials. :452-458 |
| ISSN: | 1662-7482 |
| Popis: | A cluster kernel semi-supervised support vector machine (CKS3VM) based on spectral cluster algorithm is proposed and applied in winch fault classification in this paper. The spectral clustering method is used to re-represent original data samples in an eigenvector space so as to make the data samples in the same cluster gather together much better. Then, a cluster kernel function is constructed upon the eigenvector space. Finally, a cluster kernel S3VM is designed which can satisfy the cluster assumption of semi-supervised study. The experiments on winch fault classification show that the novel approach has high classification accuracy. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|