KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion
Novelty detection is an important research issue in the field of machine learning.Till now,there exist lots of novelty detection approaches.As a commonly used kernel method,kernel principal component analysis(KPCA)has been successfully applied to deal with the problem of novelty detection.However,th...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | zho |
| Published: |
Editorial office of Computer Science
2022-08-01
|
| Series: | Jisuanji kexue |
| Subjects: | |
| Online Access: | https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2022-49-8-267.pdf |
| _version_ | 1797845116284567552 |
|---|---|
| author | LI Qi-ye, XING Hong-jie |
| author_facet | LI Qi-ye, XING Hong-jie |
| author_sort | LI Qi-ye, XING Hong-jie |
| collection | DOAJ |
| description | Novelty detection is an important research issue in the field of machine learning.Till now,there exist lots of novelty detection approaches.As a commonly used kernel method,kernel principal component analysis(KPCA)has been successfully applied to deal with the problem of novelty detection.However,the traditional KPCA based novelty detection method is very sensitive to noise.If there exist noise in the given training samples,the detection performance of KPCA based novelty detection method may be decreased.To enhance the anti-noise ability of KPCA based novelty detection method,a maximum correntropy criterion(MCC)based novelty detection method is proposed.Correntropy in information theoretic learning is utilized to substitute the <i><sub>2</sub></i>-norm based measure in KPCA based novelty detection method.By adjusting the width parameter of the correntropy function,the adverse effect of noise can be alleviated.The half-quadratic optimization technique is used to solve the optimization problem of the proposed method.The local optimal solution can thus be obtained after a few iterations.Moreover,the algorithmic description of the proposed method is provided,and the computational complexity of the corresponding algorithm is analyzed.Experimental results on the 16 UCI benchmark data sets demonstrate that the proposed method obtains better anti-noise and generalization performance in comparison with the other four related approaches. |
| first_indexed | 2024-04-09T17:33:22Z |
| format | Article |
| id | doaj.art-2fef1c5ca5ba411690414405d7746c8f |
| institution | Directory Open Access Journal |
| issn | 1002-137X |
| language | zho |
| last_indexed | 2024-04-09T17:33:22Z |
| publishDate | 2022-08-01 |
| publisher | Editorial office of Computer Science |
| record_format | Article |
| series | Jisuanji kexue |
| spelling | doaj.art-2fef1c5ca5ba411690414405d7746c8f2023-04-18T02:32:22ZzhoEditorial office of Computer ScienceJisuanji kexue1002-137X2022-08-0149826727210.11896/jsjkx.210700175KPCA Based Novelty Detection Method Using Maximum Correntropy CriterionLI Qi-ye, XING Hong-jie0Hebei Key Laboratory of Machine Learning and Computational Intelligence,College of Mathematics and Information Science,Hebei University,Baoding,Hebei 071002,ChinaNovelty detection is an important research issue in the field of machine learning.Till now,there exist lots of novelty detection approaches.As a commonly used kernel method,kernel principal component analysis(KPCA)has been successfully applied to deal with the problem of novelty detection.However,the traditional KPCA based novelty detection method is very sensitive to noise.If there exist noise in the given training samples,the detection performance of KPCA based novelty detection method may be decreased.To enhance the anti-noise ability of KPCA based novelty detection method,a maximum correntropy criterion(MCC)based novelty detection method is proposed.Correntropy in information theoretic learning is utilized to substitute the <i><sub>2</sub></i>-norm based measure in KPCA based novelty detection method.By adjusting the width parameter of the correntropy function,the adverse effect of noise can be alleviated.The half-quadratic optimization technique is used to solve the optimization problem of the proposed method.The local optimal solution can thus be obtained after a few iterations.Moreover,the algorithmic description of the proposed method is provided,and the computational complexity of the corresponding algorithm is analyzed.Experimental results on the 16 UCI benchmark data sets demonstrate that the proposed method obtains better anti-noise and generalization performance in comparison with the other four related approaches.https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2022-49-8-267.pdfkernel principal component analysis|correntropy|half-quadratic optimization|novelty detection|information theoretic learning |
| spellingShingle | LI Qi-ye, XING Hong-jie KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion Jisuanji kexue kernel principal component analysis|correntropy|half-quadratic optimization|novelty detection|information theoretic learning |
| title | KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion |
| title_full | KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion |
| title_fullStr | KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion |
| title_full_unstemmed | KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion |
| title_short | KPCA Based Novelty Detection Method Using Maximum Correntropy Criterion |
| title_sort | kpca based novelty detection method using maximum correntropy criterion |
| topic | kernel principal component analysis|correntropy|half-quadratic optimization|novelty detection|information theoretic learning |
| url | https://www.jsjkx.com/fileup/1002-137X/PDF/1002-137X-2022-49-8-267.pdf |
| work_keys_str_mv | AT liqiyexinghongjie kpcabasednoveltydetectionmethodusingmaximumcorrentropycriterion |