Abstract
Artificial immune system (AIS)-based pattern classification approach is relatively new in the field of pattern recognition. The study explores the potentiality of this paradigm in the context of prototype selection task that is primarily effective in improving the classification performance of nearest-neighbor (NN) classifier and also partially in reducing its storage and computing time requirement. The clonal selection model of immunology has been incorporated to condense the original prototype set, and performance is verified by employing the proposed technique in a practical optical character recognition (OCR) system as well as for training and testing of a set of benchmark databases available in the public domain. The effect of control parameters is analyzed and the efficiency of the method is compared with another existing techniques often used for prototype selection. In the case of the OCR system, empirical study shows that the proposed approach exhibits very good generalization ability in generating a smaller prototype library from a larger one and at the same time giving a substantial improvement in the classification accuracy of the underlying NN classifier. The improvement in performance has been statistically verified. Consideration of both OCR data and public domain datasets demonstrate that the proposed method gives results better than or at least comparable to that of some existing techniques.
Similar content being viewed by others
Notes
Instead of Hamming distance, the present experiment also considers the use of Euclidean distance in measuring stimulation value. In this case, 448-dimensional features need not be converted into binary. Since the minimum and maximum values that can occur in each dimension are known, distance between a pair of patterns is normalized to give a stimulation measure in [0, 1]. However, by using Euclidean distance instead of Hamming distance no significant change was observed in the experimental results. All the results presented here are obtained when Hamming distance was used to measure stimulation.
No iteration is needed if an antigen finds an exact match in the memory. In such a case, producing clones won’t help to find any better B cell and that is why hyper-mutation phase is not invoked at all.
References
Cover TM, Hart PE (1967) Nearest neighbor pattern classification. IEEE Trans Inform Theory 13:21–27
Hart PE (1968) The condensed nearest neighbor rule. IEEE Trans Inform Theory (IT) 14(3):515–516
Swonger CW (1972) Sample set condensation for a condensed NN decision rule for pattern recognition. In: Watanab S (ed) Frontiers of pattern recognition. Academic Press, New York, pp 511–519
Gates GW (1972) The reduced nearest neighbour rule. IEEE Trans Inform Theory 18(3):431–433
Sanchez JS, Pla F, Ferri FJ (1995) Prototype selection for the nearest neighbour rule through proximity graphs. Pattern Recognit Lett (PRL) 18(6):507–513
Skalak DB (1995) Prototype selection for composite nearest neighbor classifiers. PhD thesis, Computer Science, University of Massachusetts Amherst, USA
Wilson DR, Martinez TR (2000) Reduction techniques for instance-based learning algorithms. Mach Learn 38(3):257–286
Brighton H, Mellish C (2002) Advances in instance selection for instance-based learning algorithms. Data Min Knowl Discov 6:153–172
Susheela Devi V, Narasimha Murty M (2002) An incremental prototype set building technique. Pattern Recognit 35:505–513
Mollineda R, Ferri FJ, Vidal E (2002) An efficient prototype merging strategy for the condensed 1-NN rule through class-conditional hierarchical clustering. Pattern Recognit 35:2771–2782
Pekalska E, Duin RPW (2002) Prototype selection for finding efficient representations of dissimilarity data. In: Sixteenth international conference on pattern recognition (ICPR), vol 3, pp 37–40
Sanchez JS, Barandela R, Marques AI, Alejo R, Badenas J (2003) Analysis of new techniques to obtain quality training sets. Pattern Recognit Lett (PRL) 24(7):1015–1022
Cano JR, Herrera F, Lozano M (2003) Using evolutionary algorithms as instance selection for data reduction in KDD: an experimental study. IEEE Trans Evol Comput 7(6):561–575
Sanchez JS (2004) High training set size reduction by space partitioning and prototype abstraction. Pattern Recognit 37(7):1561–1564
Li Y, Hu Z, Cai Y, Zhang W (2005) Support vector based prototype selection method for nearest neighbor rules. Advances in natural computation. Lecture notes in computer science, vol 3610. Springer, Berlin, pp 528–535
Barandela R, Ferri FJ, Sanchez JS (2005) Decision boundary preserving prototype selection for nearest neighbor classification. Int J Pattern Recognit Artif Intell (IJPRAI) 19(6):787–806
Huang DD, Chow TWS (2006) Enhancing density-based data reduction using entropy. Neural Comput 18(2):470–495
Paredes R, Vidal E (2006) Learning prototypes and distances: a prototype reduction technique based on nearest neighbor error minimization. Pattern Recognit 39(2):180–188
Kim S-W, John B Oommen (2003) A brief taxonomy and ranking of creative prototype reduction schemes. Pattern Anal Appl (PAA) 6(3):232–244
Dasgupta D, Ji Z, Gonzalez FF (2003) Artificial immune system (AIS) research in the last five years. In: Congress on evolutionary computation (CEC’03) 1:123–130
Dasgupta D (1998) An overview of artificial immune systems and their applications. In: Dasgupta D (ed) Artificial immune systems and their applications. Springer, Berlin, pp 3–21
Zheng Tang, Koichi Tashima, Cao QP (2003) Pattern recognition system using a clonal selection-based immune network. Syst Comput Japan 34(12):56–63
Ji Z, Dasgupta D (2004) Real-valued negative selection algorithm with variable-sized detectors. In: Proceedings of GECCO. LNCS, vol 3102, pp 287–298
de Castro LN, Zuben FVJ (2002) Learning and optimization using the clonal selection principle. IEEE Trans Evol Comput Spec Issue Artif Immune Syst 6:239–251
Carter HJ (2000) The immune system as a model for pattern recognition and classification. J Am Med Inf Assoc 7(3):28–41
Watkins AB (2001) AIRS: a resource limited artificial immune classifier. Master’s dissertation, Department of Computer Science, Mississippi State University
Garain U, Chakraborty PM, Dutta Majumder D (2006) Improvement of OCR accuracy by similar character pair discrimination: an approach based on artificial immune system. In: Proceedings of the 18th international conference on pattern recognition (ICPR), August 2006, Hong Kong II, pp 1046–1049
Garain U, Chakraborty PM, Dasgupta D (2006) Recognition of handwritten indic script using clonal selection algorithm. In: Bersini H, Carneiro J (eds) 5th international conference on artificial immune systems (ICARIS), 2006, LNCS, vol 4163. Springer, Berlin, pp 256–266
de Castro LN, Timmis J (2002) Artificial immune systems: a novel approach to pattern recognition. In: Alonso L, Corchado J, Fyfe C (eds) Artificial neural networks in pattern recognition. University of Paisley, pp 67–84
Timmis J (2001) Artificial immune systems: a novel data analysis techniques inspired by the immune network theory. Ph.D. thesis, University of Wales, Aberystwyth
Burnet FM (1959) The clonal selection theory of acquired immunity. Vanderbuilt University, Nashville, TN, USA
Jerne NK (1974) Towards a network theory of the immune system. Ann Immunol (Inst Pasteur) 125C:373–389
Perelsen AS, Oster GF (1979) Theoretical studies of clonal selection: minimal antibody repertoire size and reliability of self-nonself discrimination. J Theor Biol 81:645–670
Chaudhuri BB, Garain U, Mitra M (2003) On OCR of the most popular Indian scripts: Devnagari and Bangla,” Technical report, no. TR/ISI/CVPR/03/2003, Indian Statistical Institute, Kolkata, August 2003. A product named Chitrankan is developed based on this research (http://www.cdac.in/HTML/gist/products/chitra.asp)
Baird HS (1993) Perfect metrics. In: Proceedings of the second international conference on document analysis and recognition, Tsukuba, Japan, pp 593–597
Garain U, Chaudhuri BB (1998) Compound character recognition by run number based metric distance. In: Proceedings of the IS&T/SPIE’s 10th international symposium on electronic imaging: Science & Technology, SPIE, vol 3305. San Jose, CA, USA, pp 90–97
Kohonen T (1990) The Self-organizing map. Proc IEEE 78(9):464–1480
Blake C, Keogh E, Merz C. UCI repository of machine learning databases. http://www.ics.uci.edu/mlearn/MLRepository.html
D. Statistics and M.S.S.S. University, Statlog Corp.http://ftp.strath.ac.uk
Box GEP, Hunter GW, Hunter SJ (1978) Statistics for experimenters. Wiley, New York
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Garain, U. Prototype reduction using an artificial immune model. Pattern Anal Applic 11, 353–363 (2008). https://doi.org/10.1007/s10044-008-0106-1
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10044-008-0106-1