Abstract Most biometric identification applications suffer from the curse of dimensionality as the database size becomes very large, which could negatively affect both the identification performance and speed. In this paper, we use Projection Pursuit (PP) methods to determine clusters of individuals. Support Vector Machine (SVM) classifiers are then applied on each cluster of users separately. PP clustering is conducted using Friedman and Kurtosis projection indices optimized by Genetic Algorithm and Particle Swarm Optimization methods. Experimental results obtained using YALE face database showed improvement in the performance and speed of face identification system.

References

[1] Awad M., “Forecasting of Chaotic Time Series Using RBF Neural Networks Optimized by Genetic Algorithms,” The International Arab Journal of Information Technology, vol. 14, no. 6, pp. 826-834, 2017.

[2] Belhumeur P., Hespanha J., and Kriegman D., “Eigenfaces vs. Fisherfaces: Recognition Using Class Specific Linear Projection,” IEEE Trans. on Pattern Analysis and Machine Intelligence, vol. 19, no. 7, pp. 711-720, 1997.

[3] Berro A., Marie-Sainte S., and Ruiz-Gazen A., “Genetic Algorithms and Particle Swarm Optimization for Exploratory Projection Pursuit,” Annals of Mathematics and Artificial Intelligence, vol. 60, no. 1-2, pp. 153-178, 2010.

[4] Boley D. and Cao D., “Training Support Vector Machine Using Adaptive Clustering,” in Proceedings of the 4th SIAM International Conference on Data Mining, Lake Buena Vista, pp. 126-137, 2004.

[5] Bakır G., Bottou L., and Wiston J., “Breaking SVM Complexity with Cross-Training,” in Proceedings of the 17th International Conference on Neural Information Processing Systems, Vancouver, pp. 81-88, 2004.

[6] Bridle J., Heading A., and MacKay D., “Unsupervised Classifiers, Mutual Information and Phantom Targets,” in Proceedings of the 4th International Conference on Neural Information Processing Systems, Denver, pp. 1096-1101, 1991.

[7] Burges J., “A Tutorial on Support Vector Machines for Pattern Recognition,” Data Mining and Knowledge Discovery, vol. 2, no. 2, pp. 121- 167, 1998.

[8] Chang C. and Lin C., “LIBSVM: a Library for Support Vector Machines,” ACM Transactions on Intelligent Systems and Technology, vol. 2, no. 3, pp. 1-27, 2011.

[9] Chowdhury N., Cook D., Hofmann H., Majumder M., Lee E., and Toth A., “Using Visual Statistical Inference to Better Understand Random Class Separations In High Dimension, Low Sample Size Data,” Computation Statistique, vol. 30, no. 2, pp. 293-316, 2015. 126 The International Arab Journal of Information Technology, Vol. 17, No. 1, January 2020

[10] Collobert R., Bengio S., and Bengio Y., “A Parallel Mixture of Svms for Very Large Scale Problems,” Neural Computation, vol. 14, no. 5, pp. 1105-1114, 2002.

[11] Duda R., Hart P., and Stork D., Pattern Classification, Wiley, 2001.

[12] Espezua S., Villanueva E., Maciel C., and Carvalho A., “A Projection Pursuit Framework For Supervised Dimension Reduction of High Dimensional Small Sample Datasets,” Neuro Computing Journal, vol. 149, pp. 767-776, 2015.

[13] Fisher R., “The Statistical Utilization of Multiple Measurements,” Annals of Eugenics, vol. 8, no. 4, pp. 376-386, 1938.

[14] Fogel L., Owens A., and Walsh M., Artificial Intelligence through Simulated Evolution, Wiley, 1966.

[15] Friedman J., “Exploratory Projection Pursuit,” Journal of the American Statistical Association, vol. 82, no. 1, pp. 249-266, 1987.

[16] Ghouzali S. and Marie-Sainte S., “Support Vector Machine Based Exploratory Projection Pursuit Optimization for User Face Identification,” in Proceedings of International Conference on Information and Communication Technology Research, Abu Dhabi, pp. 322-325, 2015.

[17] Grochowski M. and Duch W., “Projection Pursuit Constructive Neural Networks Based on Quality of Projected Clusters,” in Proceedings of International Conference on Artificial Neural Networks, Czech Republic, pp. 754-762, 2008.

[18] Gutfinger D. and Sklansky J., “Robust Classifiers by Mixed Adaptation,” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 13, no. 6, pp. 552-567, 1991.

[19] Holland J., Adaptation in Natural and Artificial Systems, University of Michigan Press, 1975.

[20] Hsu C. and Lin C., “A Comparison of Methods for Multi-Class Support Vector Machines,” IEEE Transactions on Neural Networks, vol. 13, no. 2, pp. 415-425, 2002.

[21] Huber P., “Projection Pursuit,” Annals of Statistics, vol. 13, no. 2, pp. 435-475, 1985.

[22] Intrator N., “Combining Exploratory Projection Pursuit and Projection Pursuit Regression with Application to Neural Networks,” Journal of Neural Computation Archive, vol. 5, no. 3, pp. 443-455, 1993.

[23] Johnson R. and Wichern D., Applied Multivariate Statistical Analysis, Prentice-Hall, 2002.

[24] Jones M. and Sibson R., “What is Projection Pursuit?,” Journal of the Royal Statistical Society, vol. 150, no.1, pp. 1-37, 1987.

[25] Kennedy J., Eberhart R., and Shi Y., Swarm Intelligence, Morgan Kauffman, 2001.

[26] Lee E., Cook D., Klinke S., and Lumley T., “Projection Pursuit for Exploratory Supervised Classification,” Journal of Computational and Graphical Statistics, vol. 14, no. 4, pp. 831-846, 2005.

[27] Lee E. and Cook D., “A Projection Pursuit Index for Large P Small N Data,” Statistical Computing, vol. 20, no. 3, pp. 318-392, 2010.

[28] Lee Y., Cook D., Park J., and Lee E., “PPtree: Projection Pursuit Classification Tree,” Electronic Journal of Statistics, vol. 7, pp. 1369- 1386, 2013.

[29] Lee Y. and Mangasarian O., “RSVM: Reduced Support Vector Machines,” in Proceedings of the 1st SIAM International Conference on Data Mining, Chicago, pp. 1-17, 2001.

[30] Marie-Sainte S., “Detection and Visualization of Non-Linear Structures in Large Datasets Using Exploratory Projection Pursuit Laboratory (EPP- Lab) Software,” Journal of King Saud University -Computer and Information Sciences, vol. 29, no. 1, pp. 2-18, 2017.

[31] Marie-Sainte S., Berro A., and Ruiz-Gazen A., “An Efficient Optimization Method for Revealing Local Optima of Projection Pursuit Indices,” in Proceedings of International Conference on Swarm Intelligence, Brussels, pp. 60-71, 2010.

[32] Mekhmoukh A. and Mokrani K., “MR Brain Image Segmentation Using an Improved Kernel Fuzzy Local Information C-Means Based Wavelet, Particle Swarm Optimization (PSO) Initialization and Outlier Rejection with Level Set Methods,” The International Arab Journal of Information Technology, vol. 15, no. 4, pp. 683- 692, 2018.

[33] Morton S., “Interpretable Projection Pursuit,” Ph.D. Dissertation, Standford University, Standford, 1989.

[34] Moujahdi C., Ghouzali S., Mikram M., Wadood A., and Rziza1M., “Inter-Communication Classification for Multi-View Face Recognition,” The International Arab Journal of Information Technology, vol. 11, no. 4, pp. 387-395, 2014.

[35] Parsopoulos K., Particle Swarm Methods, Springer International Publishing, 2015.

[36] Pavlov D., Chudova D., and Smyth P., “Towards Scalable Support Vector Machines Using Squashing,” in Proceedings of the 6th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Boston, pp. 295-299, 2000.

[37] Pavlov D., Mao J., and Dom B., “Scaling-up Support Vector Machines Using Boosting Algorithm,” in Proceedings 15th International Conference on Pattern Recognition, Barcelona, pp. 2219-2222, 2000.

[38] Peña D. and Prieto F., “Cluster Identification Using Projections,” Journal of the American Face Identification based Bio-Inspired Algorithms 127 Statistical Association, vol. 96, no. 456, pp. 1433-1445, 2001.

[39] Platt J., “Sequential Minimal Optimization: A Fast Algorithm for Training Support Vector Machines,” Technical Report MSR-TR-98-14 Microsoft Research, MSR-TR-98-14, 1998.

[40] Rechenberg I., Evolutionsstrategie: Optimierung Technischer Systeme Nach Prinzipien Der Biologischen Evolution, Frommann-Holzboog, Stuttgart, 1973.

[41] Ren Y., Liua H., Yao X., and Liu M., “Prediction of Ozone Tropospheric Degradation Rate Constants by Projection Pursuit Regression,” Analytica Chimica Acta, vol. 589, no. 1, pp. 150- 158, 2007.

[42] Schohn G. and Cohn D., “Less is more: Active Learning with Support Vector Machines,” in Proceedings of the 7th International Conference on Machine Learning, San Francisco, pp. 839- 846, 2000.

[43] Schwefel H., Numerical Optimization of Computer Models, John Wiley and Sons, 1981.

[44] Sun J., “Some Practical Aspects of Exploratory Projection Pursuit,” SIAM Journal on Scientific Computing, vol. 14, no. 1, pp. 68-80, 1993.

[45] Tong S. and Koller D., “Support Vector Machine Active Learning with Applications to Text Classification,” in Proceedings of the 17th International Conference on Machine Learning, San Francisco, pp. 999-1006, 2000.

[46] Tsang W., Kwok J., and Cheung P., “Very Large SVM Training Using Core Vector Machines,” in Proceedings of the 10th International Workshop on Artificial Intelligence and Statistics, Barbados, 2005.

[47] Vapnik V., The Nature of Statistical Learning Theory, Springer-Verlag, 1999.

[48] Yamac M., “Can We Do Better By Combining Supervised and Non-Supervised Machine Learning for Pattern Analysis,” Ph.D. Dissertation, Brown University, 1969.

[49] Yu H., Yang J., and Han J., “Classifying Large Data Sets Using SVM with Hierarchical Clusters,” in Proceedings of the 9th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Washington DC, pp. 306-315, 2003.

[50] Zhu J., Li W., Li H., Wu Q., and Zhang L., “A Novel Swarm Intelligence Algorithm for the Evacuation Routing Optimization Problem,” The International Arab Journal of Information Technology, vol. 14, no. 6, pp. 880-889, 2017. Sanaa Ghouzali received both the Master's and the Ph.D. degrees in Computer Sciences from University Mohamed V-Agdal, Rabat, Morocco, in 2004 and 2009, respectively. In 2005 she has received a Fulbright grant as a visiting student at the Visual and Communication Laboratory of Cornell University, Ithaca, NY, USA. Between 2009 and 2011, she was an Assistant Professor at ENSA (the National school of Applied Sciences) with in Abdelmalek Essaadi University. Starting 2012, she joined King Saud University in the College of Computer and Information Sciences. Her research interests include statistical pattern detection and recognition, Biometrics, Biometric Security and Protection. Souad Larabi is an assistant professor at the department of Computer Science at Prince Sultan University. She was an assistant professor at Information Technology department at King Saud University from 2012 to 2016. She worked as an associate researcher in the department of Computer Science at Toulouse1 Capitole University, France. She earned her PhD from same university in the area of bio-inspired algorithms in June 2011. She holds an M.S.c degree in Mathematics, Computing, Decision and Organization at Sorbonne Paris1 University, France. Souad is graduated from USTHB University in Algeria in the area of Operational Research. Her research interests include Combinatorial Optimization, Meta-heuristics, Artificial Intelligent, Bioinformatics, Data Mining and Biometric identification.