Li Xiang
Faculty of Computer Engineering, Huaiyin Institute of Technology, Huai�an 223003, China
Zhu Quanyin
Faculty of Computer Engineering, Huaiyin Institute of Technology, Huai�an 223003, China
Wang Zun
School of Electronic Engineering and Optoelectronic Technology, Nanjing University of Science and Technology, Nanjing 210094, China
ABSTRACT
In view that the traditional Learning Vector Quantization (LVQ) neural network is sensitive to initial value and has a lower stability of the algorithm, a new LVQ model with the AdaBoost algorithm was put forward to improve the forecasting accuracy and generalization ability. Firstly, the method performed the pre-treatment for the historical data and initialized the distribution weights of test data. Secondly, it selected different hidden layer nodes and network learning functions to construct weak predictors of LVQ neural network and trained the sample data repeatedly. At last, it made more weak predictors of LVQ neural network to form a new strong predictor by AdaBoost algorithm for classification. A simulation experiment for the 6 data sets of UCI was carried out. The results show that this method has improved the classification accuracy nearly 5% compared to the traditional LVQ neural network and has a better stability of the algorithm. This method provides references for the LVQ neural network.
PDF References Citation
How to cite this article
Li Xiang, Zhu Quanyin and Wang Zun, 2013. Research of Improved LVQ Neural Network by AdaBoost Algorithm. Journal of Applied Sciences, 13: 2658-2663.
DOI: 10.3923/jas.2013.2658.2663
URL: https://scialert.net/abstract/?doi=jas.2013.2658.2663
DOI: 10.3923/jas.2013.2658.2663
URL: https://scialert.net/abstract/?doi=jas.2013.2658.2663
REFERENCES
- Chang, C.Y. and D.F. Zhuang, 2007. A fuzzy-based learning vector quantization neural network for recurrent nasal papilloma detection. IEEE Trans. Circuits Syst., 54: 2619-2626.
CrossRef - Filippi, A.M. and J.R. Jensen, 2007. Effect of continuum removal on hyperspectral coastal vegetation classification using a fuzzy learning vector quantizer. IEEE Trans. Geosci. Remote Sensing, 45: 1857-1869.
CrossRef - Gauri, S.K., 2010. Control chart pattern recognition using feature-based learning vector quantization. Int. J. Adv. Manuf. Technol., 48: 1061-1073.
CrossRefDirect Link - Guo, P., Z. Jiang, S. Lin and Y. Yao, 2012. Combining LVQ with SVM technique for image semantic annotation. Neural Comput. Appl., 21: 735-746.
CrossRefDirect Link - Hammer, B. and T. Villmann, 2002. Generalized relevance learning vector quantization. Neural Networks, 15: 1059-1068.
Direct Link - Jiang, D. and C. Liu, 2011. Machine condition classification using deterioration feature extraction and anomaly determination. IEEE Trans. Reliab., 60: 41-48.
CrossRef - Li, X. and Q.Y. Zhu, 2012. Tax forecasting based on Adaboost algorithm and BP neural network. J. Comput. Appl., 32: 3558-3560.
Direct Link - Biehl, M., 2012. Admire LVQ-adaptive distance measures in relevance learning vector quantization. Kunstliche Intelligenz, 26: 391-395.
CrossRefDirect Link - Biehl, M., A. Ghosh, B. Hammer and Y. Bengio, 2007. Dynamics and generalization ability of LVQ algorithms. J. Mach. Learn. Res., 8: 323-360.
Direct Link - Boubaker, M., M. Akil, K.B. Khalifa, T. Grandpierre and M.H. Bedoui, 2010. Implementation of an LVQ neural network with a variable size: Algorithmic specification, architectural exploration and optimized implementation on FPGA devices. Neural Comput. Appl., 19: 283-297.
CrossRefDirect Link - Pan, H., B. Chen and Q.W. Li, 2011. Paper currency number recognition based on binary tree and adaboost algorithm. J. Comput. Appl., 31: 396-398.
CrossRefDirect Link - Qiao, H., Y.Z. Zhou and S. Nan, 2012. Software reliability prediction based on learning vector quantization neutral network. J. Comput. Appl., 32: 1436-1438.
CrossRefDirect Link - Schneider, P., K. Bunte, H. Stiekema, B. Hammer, T. Villmann and M. Biehl, 2010. Regularization in matrix relevance learning. IEEE Trans. Neural Networks, 21: 831-840.
CrossRef - Triguero, I., J. Derrac, S. Garcia and F. Herrera, 2012. A taxonomy and experimental study on prototype generation for nearest neighbor classification. IEEE Trans. Syst. Man Cybernet. C: Applic. Rew., 42: 86-100.
CrossRef - Xu, Y., F. Shen and J. Zhao, 2012. An incremental learning vector quantization algorithm for pattern classification. Neural Comput. Appl., 21: 1205-1215.
CrossRefDirect Link