Probabilistic novelty detection with support vector machines
Novelty detection, or one-class classification, is of particular use in the analysis of high-integrity systems, in which examples of failure are rare in comparison with the number of examples of stable behaviour, such that a conventional multi-class classification approach cannot be taken. Support V...
| Main Authors: | , , , , , |
|---|---|
| Format: | Journal article |
| Published: |
Institute of Electrical and Electronics Engineers
2014
|
| _version_ | 1797063959232118784 |
|---|---|
| author | Clifton, L Clifton, D Zhang, Y Watkinson, P Tarassenko, L Yin, H |
| author_facet | Clifton, L Clifton, D Zhang, Y Watkinson, P Tarassenko, L Yin, H |
| author_sort | Clifton, L |
| collection | OXFORD |
| description | Novelty detection, or one-class classification, is of particular use in the analysis of high-integrity systems, in which examples of failure are rare in comparison with the number of examples of stable behaviour, such that a conventional multi-class classification approach cannot be taken. Support Vector Machines (SVMs) are a popular means of performing novelty detection, and it is conventional practice to use a train-validate-test approach, often involving cross-validation, to train the one-class SVM, and then select appropriate values for its parameters. An alternative method, used with multi-class SVMs, is to calibrate the SVM output into conditional class probabilities. A probabilistic approach offers many advantages over the conventional method, including the facility to select automatically a probabilistic novelty threshold. The contributions of this paper are (i) the development of a probabilistic calibration technique for one-class SVMs, such that on-line novelty detection may be performed in a probabilistic manner; and (ii) the demonstration of the advantages of the proposed method (in comparison to the conventional one-class SVM methodology) using case studies, in which one-class probabilistic SVMs are used to perform condition monitoring of a high-integrity industrial combustion plant, and in detecting deterioration in patient physiological condition during patient vital-sign monitoring. © 2014 IEEE. |
| first_indexed | 2024-03-06T21:07:25Z |
| format | Journal article |
| id | oxford-uuid:3cf37979-a4ac-4b47-8ad1-501f63acc8a0 |
| institution | University of Oxford |
| last_indexed | 2024-03-06T21:07:25Z |
| publishDate | 2014 |
| publisher | Institute of Electrical and Electronics Engineers |
| record_format | dspace |
| spelling | oxford-uuid:3cf37979-a4ac-4b47-8ad1-501f63acc8a02022-03-26T14:16:36ZProbabilistic novelty detection with support vector machinesJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:3cf37979-a4ac-4b47-8ad1-501f63acc8a0Symplectic Elements at OxfordInstitute of Electrical and Electronics Engineers2014Clifton, LClifton, DZhang, YWatkinson, PTarassenko, LYin, HNovelty detection, or one-class classification, is of particular use in the analysis of high-integrity systems, in which examples of failure are rare in comparison with the number of examples of stable behaviour, such that a conventional multi-class classification approach cannot be taken. Support Vector Machines (SVMs) are a popular means of performing novelty detection, and it is conventional practice to use a train-validate-test approach, often involving cross-validation, to train the one-class SVM, and then select appropriate values for its parameters. An alternative method, used with multi-class SVMs, is to calibrate the SVM output into conditional class probabilities. A probabilistic approach offers many advantages over the conventional method, including the facility to select automatically a probabilistic novelty threshold. The contributions of this paper are (i) the development of a probabilistic calibration technique for one-class SVMs, such that on-line novelty detection may be performed in a probabilistic manner; and (ii) the demonstration of the advantages of the proposed method (in comparison to the conventional one-class SVM methodology) using case studies, in which one-class probabilistic SVMs are used to perform condition monitoring of a high-integrity industrial combustion plant, and in detecting deterioration in patient physiological condition during patient vital-sign monitoring. © 2014 IEEE. |
| spellingShingle | Clifton, L Clifton, D Zhang, Y Watkinson, P Tarassenko, L Yin, H Probabilistic novelty detection with support vector machines |
| title | Probabilistic novelty detection with support vector machines |
| title_full | Probabilistic novelty detection with support vector machines |
| title_fullStr | Probabilistic novelty detection with support vector machines |
| title_full_unstemmed | Probabilistic novelty detection with support vector machines |
| title_short | Probabilistic novelty detection with support vector machines |
| title_sort | probabilistic novelty detection with support vector machines |
| work_keys_str_mv | AT cliftonl probabilisticnoveltydetectionwithsupportvectormachines AT cliftond probabilisticnoveltydetectionwithsupportvectormachines AT zhangy probabilisticnoveltydetectionwithsupportvectormachines AT watkinsonp probabilisticnoveltydetectionwithsupportvectormachines AT tarassenkol probabilisticnoveltydetectionwithsupportvectormachines AT yinh probabilisticnoveltydetectionwithsupportvectormachines |