Improving learning vector quantization using data reduction

(1) Pande Nyoman Ariyuda Semadi Mail (Universitas Gadjah Mada, Indonesia)
(2) * Reza Pulungan Mail (Universitas Gadjah Mada, Indonesia)
*corresponding author


Learning Vector Quantization (LVQ) is a supervised learning algorithm commonly used for statistical classification and pattern recognition. The competitive layer in LVQ studies the input vectors and classifies them into the correct classes. The amount of data involved in the learning process can be reduced by using data reduction methods. In this paper, we propose a data reduction method that uses geometrical proximity of the data. The basic idea is to drop sets of data that have many similarities and keep one representation for each set. By certain adjustments, the data reduction methods can decrease the amount of data involved in the learning process while still maintain the existing accuracy. The amount of data involved in the learning process can be reduced down to 33.22% for the abalone dataset and 55.02% for the bank marketing dataset, respectively.


Learning vector quantization; Data reduction; Geometric proximity; Euclidean distance



Article metrics

Abstract views : 1396 | PDF views : 241




Full Text



[1] T. Kohonen, Self-Organizing Maps, 2001, vol. 30, doi: 10.1007/978-3-642-56927-2.

[2] K. M. O. Nahar, M. Abu Shquier, W. G. Al-Khatib, H. Al-Muhtaseb, and M. Elshafei, “Arabic phonemes recognition using hybrid LVQ/HMM model for continuous speech recognition,” Int. J. Speech Technol., vol. 19, no. 3, pp. 495–508, Sep. 2016, doi: 10.1007/s10772-016-9337-5.

[3] F. Camastra, “Handwritten Greek Character Recognition with Learning Vector Quantization,” in: Apolloni B., Howlett R.J., Jain L. (eds) International Conference on Knowledge-Based and Intelligent Information and Engineering Systems (KES 2007), Lecture Notes in Computer Science, vol. 4694, pp. 267–274, doi: 10.1007/978-3-540-74829-8_33.

[4] Chuanfeng Lv, Xing An, Zhiwen Liu, and Qiangfu Zhao, “Dual Weight Learning Vector Quantization,” in 2008 9th International Conference on Signal Processing, 2008, pp. 1722–1725, doi: 10.1109/ICOSP.2008.4697470.

[5] J. C. Yang, S. Yoon, and D. S. Park, “Applying Learning Vector Quantization Neural Network for Fingerprint Matching,” in: Sattar A., Kang B. (eds) Australasian Joint Conference on Artificial Intelligence (AI 2006), Lecture Notes in Computer Science, vol. 4304, 2006, pp. 500–509, doi: 10.1007/11941439_54.

[6] C. A. de Luna-Ortega, J. A. Ramirez-Marquez, M. Mora-Gonzalez, J. C. Martinez-Romo, and C. A. Lopez-Luevano, “Fingerprint Verification Using the Center of Mass and Learning Vector Quantization,” in 2013 12th Mexican International Conference on Artificial Intelligence, 2013, pp. 123–127, doi: 10.1109/MICAI.2013.21.

[7] R. J. T. Morris, L. D. Rubin, and H. Tirri, “Neural network techniques for object orientation detection. Solution by optimal feedforward network and learning vector quantization approaches,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 12, no. 11, pp. 1107–1115, 1990, doi: 10.1109/34.61712.

[8] J. Cannady, “Distributed Detection of Attacks in Mobile Ad Hoc Networks Using Learning Vector Quantization,” in 2009 Third International Conference on Network and System Security, 2009, pp. 571–574, doi: 10.1109/NSS.2009.99.

[9] J. Jing, J. Wang, P. Li, and Y. Li, “Automatic Classification of Woven Fabric Structure by Using Learning Vector Quantization,” Procedia Eng., vol. 15, pp. 5005–5009, 2011, doi: 10.1016/j.proeng.2011.08.930.

[10] J. C. Ortiz-Bayliss, H. Terashima-Marín, and S. E. Conant-Pablos, “Learning vector quantization for variable ordering in constraint satisfaction problems,” Pattern Recognit. Lett., vol. 34, no. 4, pp. 423–432, Mar. 2013, doi: 10.1016/j.patrec.2012.09.009.

[11] K. Song et al., “Multi-mode energy management strategy for fuel cell electric vehicles based on driving pattern identification using learning vector quantization neural network algorithm,” J. Power Sources, vol. 389, pp. 230–239, Jun. 2018, doi: 10.1016/j.jpowsour.2018.04.024.

[12] M. A. Syufagi, M. Hariadi, and M. H. Purnomo, “A Cognitive Skill Classification Based on Multi Objective Optimization Using Learning Vector Quantization for Serious Games,” ITB J. Inf. Commun. Technol., vol. 5, no. 3, pp. 189–206, 2011, doi: 10.5614/itbj.ict.2011.5.3.3.

[13] M. F. Umer and M. S. H. Khiyal, “Classification of Textual Documents Using Learning Vector Quantization,” Inf. Technol. J., vol. 6, no. 1, pp. 154–159, Jan. 2007, doi: 10.3923/itj.2007.154.159.

[14] M. Strickert, T. Bojer, and B. Hammer, “Generalized Relevance LVQ for Time Series,” in: Dorffner G., Bischof H., Hornik K. (eds) International Conference on Artificial Neural Networks (ICANN 2001), Lecture Notes in Computer Science, vol. 2130, 2001, pp. 677–683, doi: 10.1007/3-540-44668-0_94.

[15] L. Sa Silva, A. Ferrari Dos Santos, A. Montes, and J. Da Silva Simoes, “Hamming Net and LVQ Neural Networks for Classification of Computer Network Attacks: A Comparative Analysis,” in 2006 Ninth Brazilian Symposium on Neural Networks (SBRN’06), 2006, pp. 13–13, doi: 10.1109/SBRN.2006.21.

[16] C.-R. Chen, L.-T. Tsai, and C.-C. Yang, “Supervised learning vector quantization for projecting missing weights of hierarchical neural networks,” WSEAS Trans. Inf. Sci. Appl., vol. 7, no. 6, pp. 799–808, 2010, available at : Google Scholar.

[17] N. Kitajima, “A new method for initializing reference vectors in LVQ,” in Proceedings of ICNN’95 - International Conference on Neural Networks, 1995, vol. 5, pp. 2775–2779 vol.5, doi: 10.1109/ICNN.1995.488170.

[18] M. Blachnik and W. Duch, “LVQ algorithm with instance weighting for generation of prototype-based rules,” Neural Networks, vol. 24, no. 8, pp. 824–830, Oct. 2011, doi: 10.1016/j.neunet.2011.05.013.

[19] M.-T. Vakil-Baghmisheh and N. Pavešić, “Premature clustering phenomenon and new training algorithms for LVQ,” Pattern Recognit., vol. 36, no. 8, pp. 1901–1912, Aug. 2003, doi: 10.1016/S0031-3203(02)00291-1.

[20] D. R. Wilson and T. R. Martinez, “Reduction Techniques for Instance-Based Learning Algorithms,” Mach. Learn., vol. 38, no. 3, pp. 257–286, Mar. 2000, doi: 10.1023/A:1007626913721.

[21] D. P. Ismi, S. Panchoo, and M. Murinto, “K-means clustering based filter feature selection on high dimensional data,” Int. J. Adv. Intell. Informatics, vol. 2, no. 1, pp. 38–45, Mar. 2016, doi: 10.26555/ijain.v2i1.54.

[22] D. R. Wilson and T. R. Martinez, “Instance pruning techniques,” in Proceedings of the Fourteenth International Conference on Machine Learning (ICML 1997), 1997, vol. 97, no. 1997, pp. 400–411, available at : Google Scholar.

[23] C. E. Pedreira, “Learning vector quantization with training data selection,” IEEE Trans. Pattern Anal. Mach. Intell., vol. 28, no. 1, pp. 157–162, Jan. 2006, doi: 10.1109/TPAMI.2006.14.

[24] Chien-Hsing Chou, Bo-Han Kuo, and Fu Chang, “The Generalized Condensed Nearest Neighbor Rule as A Data Reduction Method,” in 18th International Conference on Pattern Recognition (ICPR’06), 2006, vol. 2, pp. 556–559, doi: 10.1109/ICPR.2006.1119.

[25] J. Wang, S. Yue, X. Yu, and Y. Wang, “An efficient data reduction method and its application to cluster analysis,” Neurocomputing, vol. 238, pp. 234–244, May 2017, doi: 10.1016/j.neucom.2017.01.059.

[26] S. Ougiaroglou, K. I. Diamantaras, and G. Evangelidis, “Exploring the effect of data reduction on Neural Network and Support Vector Machine classification,” Neurocomputing, vol. 280, pp. 101–110, Mar. 2018, doi: 10.1016/j.neucom.2017.08.076.

[27] L. Fausett, Ed., Fundamentals of Neural Networks: Architectures, Algorithms, and Applications. Upper Saddle River, NJ, USA: Prentice-Hall, Inc., 1994, available at:

[28] A. Apriliani, R. Kusumaningrum, S. N. Endah, and Y. Prasetyo, “Suitability analysis of rice varieties using learning vector quantization and remote sensing images,” TELKOMNIKA (Telecommunication Comput. Electron. Control., vol. 17, no. 3, p. 1290, Jun. 2019, doi: 10.12928/telkomnika.v17i3.12234.

[29] G. Kumar, S. Sharma, and H. Malik, “Learning Vector Quantization Neural Network Based External Fault Diagnosis Model for Three Phase Induction Motor Using Current Signature Analysis,” Procedia Comput. Sci., vol. 93, pp. 1010–1016, 2016, doi: 10.1016/j.procs.2016.07.304.

[30] W. H. Nugroho, S. Handoyo, and Y. J. Akri, “An Influence of Measurement Scale of Predictor Variable on Logistic Regression Modeling and Learning Vector Quntization Modeling for Object Classification,” Int. J. Electr. Comput. Eng., vol. 8, no. 1, p. 333, Feb. 2018, doi: 10.11591/ijece.v8i1.pp333-343.

[31] H. Hartono, O. S. Sitompul, T. Tulus, and E. B. Nababan, “Biased support vector machine and weighted-smote in handling class imbalance problem,” Int. J. Adv. Intell. Informatics, vol. 4, no. 1, p. 21, Mar. 2018, doi: 10.26555/ijain.v4i1.146.

[32] Z. P. Agusta and A. Adiwijaya, “Modified balanced random forest for improving imbalanced data prediction,” Int. J. Adv. Intell. Informatics, vol. 5, no. 1, p. 58, Dec. 2018, doi: 10.26555/ijain.v5i1.255.

[33] Huan Liu and Lei Yu, “Toward integrating feature selection algorithms for classification and clustering,” IEEE Trans. Knowl. Data Eng., vol. 17, no. 4, pp. 491–502, Apr. 2005, doi: 10.1109/TKDE.2005.66.

[34] X.-B. Li, “Data reduction via adaptive sampling,” Commun. Inf. Syst., vol. 2, no. 1, pp. 53–68, 2002, doi: 10.4310/CIS.2002.v2.n1.a3.

[35] P. N. A. Semadi, “Pengembangan metode reduksi data dan inisialisasi vektor referensi pada algoritma learning vector quantization [Development of data reduction method and reference vector initialization on learning vector quantization algorithm],” Master’s Thesis, Universitas Gadjah Mada, 2016, [in Bahasa Indonesia], available at: Google Scholar.

Creative Commons License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.

International Journal of Advances in Intelligent Informatics
ISSN 2442-6571  (print) | 2548-3161 (online)
Organized by Informatics Department - Universitas Ahmad Dahlan, and ASCEE Computer Society
Published by Universitas Ahmad Dahlan
E: (paper handling issues), (publication issues)

View IJAIN Stats

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0