Kernel distance-based robust support vector method and its application in developing a robust K-chart. (English) Zbl 1095.62137
Summary: Traditional statistical process control (SPC) techniques are not applicable in many process industries due to autocorrelation among data. In addition, most conventional charts are based on the assumption that quality characteristics follow a multivariate normality assumption. Therefore, the reduction in process variability obtained through the use of SPC techniques has not been realized in the industries. However, this may not be reasonable in many real-world problems and its extension poses serious limitations. Hence, it is not only desirable, but also inevitable to have some techniques that can serve the same purpose as SPC control charts used for correlated parameters.
In this paper, a robust support vector method drawn from statistical learning theory was applied to develop a multivariate control chart based on kernel distance, which is a measure of the distance between the centre of a class and the sample to be monitored. The proposed robust chart takes advantage of information extracted from in-control preliminary samples. A robust support vector method-based chart aims to solve the over fitting problems when outliers exist in the training data set. The robust support vector method makes the decision function less sensitive towards the noise and outliers. The performance of the robust chart is tested on the problem taken from the literature and the results verify the effectiveness of the chart and validate that the robust chart is better than the conventional charts when the distribution of the quality characteristics is not multivariate normal. Experiments for the problem undertaken confirm the reduction in the number of support vectors and there is significant improvement in performance when compared with the standard support vector methods.
In this paper, a robust support vector method drawn from statistical learning theory was applied to develop a multivariate control chart based on kernel distance, which is a measure of the distance between the centre of a class and the sample to be monitored. The proposed robust chart takes advantage of information extracted from in-control preliminary samples. A robust support vector method-based chart aims to solve the over fitting problems when outliers exist in the training data set. The robust support vector method makes the decision function less sensitive towards the noise and outliers. The performance of the robust chart is tested on the problem taken from the literature and the results verify the effectiveness of the chart and validate that the robust chart is better than the conventional charts when the distribution of the quality characteristics is not multivariate normal. Experiments for the problem undertaken confirm the reduction in the number of support vectors and there is significant improvement in performance when compared with the standard support vector methods.
MSC:
| 62P30 | Applications of statistics in engineering and industry; control charts |
| 90B30 | Production models |
| 68T05 | Learning and adaptive systems in artificial intelligence |
Keywords:
support vector method; robust support vector method; kernel functions; control charts; hypersphere; principal component; regularization parameterReferences:
| [1] | DOI: 10.2307/1391421 · doi:10.2307/1391421 |
| [2] | Boser BE, Proceeding of the Fifth Annual Workshop on Computational Learning Theory (1992) |
| [3] | Box GEP, Statistical Control by Monitoring and Feedback Adjustment (1997) |
| [4] | DOI: 10.1023/A:1009715923555 · doi:10.1023/A:1009715923555 |
| [5] | Chakraborti S, J. Qual. Tech. 3 pp 304– (2001) |
| [6] | Cook DF, IIE Trans. 30 pp 227– (1998) |
| [7] | Gile L, Proceeding of the 1997 IEEE Workshop on Neural Network Signal Processing VII pp 276– (1997) |
| [8] | DOI: 10.1002/cjce.5450690106 · doi:10.1002/cjce.5450690106 |
| [9] | Haykin S, Neural Networks – A Comprehensive Foundation (1999) |
| [10] | Herbrich, R and Weston, J. 1999. Adaptive martin support vector machines for classification. Proceedings to the 9th ICANN. 1999. pp.880–895. 2 |
| [11] | DOI: 10.1080/07408179508936797 · doi:10.1080/07408179508936797 |
| [12] | DOI: 10.1016/0967-0661(95)00014-L · doi:10.1016/0967-0661(95)00014-L |
| [13] | Montgomery DC, Introduction to Statistical Quality Control (2001) |
| [14] | DOI: 10.1109/72.914517 · doi:10.1109/72.914517 |
| [15] | DOI: 10.1002/aic.690420412 · doi:10.1002/aic.690420412 |
| [16] | DOI: 10.1080/03610929508831487 · Zbl 0825.62255 · doi:10.1080/03610929508831487 |
| [17] | Saitoh, S. 1988. ”Theory of reproducing Kernels and its applications”. Harlow: Longman. · Zbl 0652.30003 |
| [18] | Schilling EG, J. Qual. Tech. 8 pp 183– (1976) |
| [19] | DOI: 10.1109/72.788641 · doi:10.1109/72.788641 |
| [20] | DOI: 10.1109/TSMCC.2002.807277 · doi:10.1109/TSMCC.2002.807277 |
| [21] | Sun, RX. 2000. Evolution computation and intelligent diagnosis. PhD dissertation, Xi an Jiatong University. 2000, China. |
| [22] | DOI: 10.1080/1352816031000075224 · Zbl 1044.62125 · doi:10.1080/1352816031000075224 |
| [23] | Vapnik V, The Nature of Statistical Learning Theory (1995) · Zbl 0833.62008 · doi:10.1007/978-1-4757-2440-0 |
| [24] | Zhang, HG. 1999. Using class centre vectors to build support vector machines. IEEE Proceedings of Neural Net Signal Processing IX. 1999. |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
