A Recognition Method Using SA Optimized SVM-DS

Abstract:

Through properly setting the simulated annealing options of acceptance function, annealing function and temperature function, an adaptive hyper-parameter estimation method using simulated annealing algorithm is applied to improve the accuracy and efficiency of SVM. While, in order to eliminate the effects of error accumulation in multi-SVM, D-S theory is employed for decision fusion of SVM classifiers. When delimiting the belief and plausibility measures, recognition capability of SVM classifiers has been taken into account. And the Dempster decision rule also has been considered to the recognition result of each SVM classifier in the fusion algorithm. Finely, with the data set in the database of Statlog for the study, the experiment result indicates that this method can significantly increase the classification accuracy and demonstrate a good performance of robust.

• Periodical: Applied Mechanics and Materials (Volumes 182-183)
• Pages: 798-804
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.182-183.798
• Online since: June 2012
• Authors: Jun Li, Hai Bo Pu
• Keywords: D-S Theory, Optimise, Recognition, Simulated Annealing (SA), Support Vector Machine (SVM)

References

[1] Arthur Zimek, Fabian Buchwald, Eibe Frank, A Study of Hierarchical and Flat Classification of Proteins, IEEE Transactions on Computational Biology and Bioinformatics, vol. 7, no. 3, pp.563-571, september (2010).

DOI: 10.1109/tcbb.2008.104

[2] V.N. Vapnik, The Nature of Statistical Learning Theory, Springer, New York, (1995).

[3] L.J. Cao, F.E.H. Tay, Support vector machine with adaptive parameters in financial time series forecasting, IEEE Trans. Neural Netw. vol. 14, no. 6, p.1506–1518, (2003).

DOI: 10.1109/tnn.2003.820556

[4] S.V.T. Ramesh, S.P. Simonovic, Optimal operation of reservoir systems using simulated annealing, Water Resources Management, vol. 16, no. 5, p.401–428, October (2002).

[5] Euan W. McGookin, David J. Murray-Smith. Submarine manoeuvring controllers' optimisation using simulated annealing and genetic algorithms, Control Engineering Practice, vol. 14, no. 1, pp.1-15, 14 January, (2006).

DOI: 10.1016/j.conengprac.2005.01.002

[6] Shih-Wei Lin, Zne-Jung Lee, Shih-Chieh Chen, Parameter determination of support vector machine and feature selection using simulated annealing approach, Applied Soft Computing, vol. 8, no. 4, p.1505–1512, September (2008).

DOI: 10.1016/j.asoc.2007.10.012

[7] T. Inagaki, Interdependence between safety-control policy and multiple-sensor schemes via Dempster–Shafer theory, , IEEE Trans. Reliability, vol. 40, no. 2, p.182–188, June (1991).

DOI: 10.1109/24.87125

[8] L. D´az-Ma´s, R. Mun˜oz-Salinas, F.J. Madrid-Cuevas, Shape from silhouette using Dempster–Shafer theory, Pattern Recognition, vol. 43, no. 6 pp.2119-2131, June (2010).

DOI: 10.1016/j.patcog.2010.01.001

[9] Chao Fu, Shan-Lin Yang, Analyzing the applicability of Dempster's rule to the combination of interval-valued belief structures, Expert Systems with Applications, vol. 38, no. 4, p.4291–4301, April 2011. in press.

DOI: 10.1016/j.eswa.2010.09.097

[10] Jun Li, Zhiyu You, An Improved Adaptive Support Vector Machine Algorithm with Combinational Fuzzy C-means Clustering, 2010 2nd International Conference on Advanced Computer Control (ICACC), p.269 – 272, March (2010).

DOI: 10.1109/icacc.2010.5486622