New Mixed Kernel Functions of SVM Used in Pattern Recognition
The pattern analysis technology based on kernel methods is a new technology, which combines good performance and strict theory. With support vector machine, pattern analysis is easy and fast. But the existing kernel function fits the requirement. In the paper, we explore the new mixed kernel functio...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Sciendo
2016-10-01
|
| Series: | Cybernetics and Information Technologies |
| Subjects: | |
| Online Access: | https://doi.org/10.1515/cait-2016-0047 |
| _version_ | 1818307293143367680 |
|---|---|
| author | Huanrui Hao |
| author_facet | Huanrui Hao |
| author_sort | Huanrui Hao |
| collection | DOAJ |
| description | The pattern analysis technology based on kernel methods is a new technology, which combines good performance and strict theory. With support vector machine, pattern analysis is easy and fast. But the existing kernel function fits the requirement. In the paper, we explore the new mixed kernel functions which are mixed with Gaussian and Wavelet function, Gaussian and Polynomial kernel function. With the new mixed kernel functions, we check different parameters. The results shows that the new mixed kernel functions have good time efficiency and accuracy. In image recognition we used SVM with two mixed kernel functions, the mixed kernel function of Gaussian and Wavelet function are suitable for more states. |
| first_indexed | 2024-12-13T06:56:04Z |
| format | Article |
| id | doaj.art-53698aae8779493788667fa24700a586 |
| institution | Directory Open Access Journal |
| issn | 1314-4081 |
| language | English |
| last_indexed | 2024-12-13T06:56:04Z |
| publishDate | 2016-10-01 |
| publisher | Sciendo |
| record_format | Article |
| series | Cybernetics and Information Technologies |
| spelling | doaj.art-53698aae8779493788667fa24700a5862022-12-21T23:56:02ZengSciendoCybernetics and Information Technologies1314-40812016-10-0116551410.1515/cait-2016-0047New Mixed Kernel Functions of SVM Used in Pattern RecognitionHuanrui Hao0Harbin University of Science and Technology, China ChinaThe pattern analysis technology based on kernel methods is a new technology, which combines good performance and strict theory. With support vector machine, pattern analysis is easy and fast. But the existing kernel function fits the requirement. In the paper, we explore the new mixed kernel functions which are mixed with Gaussian and Wavelet function, Gaussian and Polynomial kernel function. With the new mixed kernel functions, we check different parameters. The results shows that the new mixed kernel functions have good time efficiency and accuracy. In image recognition we used SVM with two mixed kernel functions, the mixed kernel function of Gaussian and Wavelet function are suitable for more states.https://doi.org/10.1515/cait-2016-0047support vector machinekernel functionspattern recognitionwavelet function |
| spellingShingle | Huanrui Hao New Mixed Kernel Functions of SVM Used in Pattern Recognition Cybernetics and Information Technologies support vector machine kernel functions pattern recognition wavelet function |
| title | New Mixed Kernel Functions of SVM Used in Pattern Recognition |
| title_full | New Mixed Kernel Functions of SVM Used in Pattern Recognition |
| title_fullStr | New Mixed Kernel Functions of SVM Used in Pattern Recognition |
| title_full_unstemmed | New Mixed Kernel Functions of SVM Used in Pattern Recognition |
| title_short | New Mixed Kernel Functions of SVM Used in Pattern Recognition |
| title_sort | new mixed kernel functions of svm used in pattern recognition |
| topic | support vector machine kernel functions pattern recognition wavelet function |
| url | https://doi.org/10.1515/cait-2016-0047 |
| work_keys_str_mv | AT huanruihao newmixedkernelfunctionsofsvmusedinpatternrecognition |