Abstract
The Fuzzy min-max network (FMN), fuzzy hypersphere neural network classifier (FHSNN), and other FMN variants are sensitive to hyperbox/hypersphere (HB/HS) size or expansion parameter. The neural network of these models needs to be repeatedly trained to find the best value of expansion parameter for which network provides near about full classification accuracy with a minimum number of HBs/HSs. The user-defined expansion parameter ranges between 0 to 1. Hence these models scan input patterns multiple times for various values of the expansion parameter. In addition to it, its value needs to be retuned by retraining the existing patterns while adapting the new input patterns or classes.
The paper proposes an Optimized Fuzzy Hypersphere Neural Network Classifier with online adaptation capability (OFHSNNwOC). The designed algorithm trains the network in a single pass that means without scanning the input dataset multiple times. The modifications proposed in the model are eviction of the expansion parameter, calculation of centre points and radii of FHSs in an optimized way, membership functions for various kinds of hyperspheres, and online capability for adaptation of new input patterns or classes in the already built model with no need for retraining.
The classifier is assessed and examined utilizing the benchmark datasets.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
McCloskey, M.,Cohen, N.J.: Catastrophic interference in connectionist networks: the sequential learning problem. In: Bower, G.H. (ed.) The Psychology of Learning and Motivation, pp. 109–165. Academic, New York (1989)
Simpson, P.K.: Fuzzy min-max neural network Part I: classification. IEEE Trans. Neural Netw. 3, 776–786 (1992)
Simpson, P.K.: Fuzzy min-max neural networks part 2: clustering. IEEE Trans. Fuzzy Syst. 1(1), 32–45 (1993)
Grossberg, S.: Adaptive pattern classification and universal recording: I. parallel development and coding of neural feature detectors. Biol. Cybern. 23(3), 121–134 (1976)
Grossberg, S.: Adaptive pattern classification and universal recording: II. feedback, expectation, olfaction, illusions. Biol. Cybern. 23(4), 187–202 (1976)
Carpenter, G.A., Grossberg, S.: A massively parallel architecture for a self-organizing neural pattern recognition machine. Comput. Vis. Graph. Image Process. 37(1), 54–115 (1987)
Mohammed, M.F., Lim, C.P.: An enhanced fuzzy min-max neural network for pattern classification. IEEE Trans. Neural Netw. Learn. Syst. 26(3), 417–429 (2015)
Kulkarni, U.V., Sontakke T.R.: Fuzzy hypersphere neural network classifier. In: 10th International Conference on Fuzzy Systems, Melbourne, Victoria, Australia, pp. 1559–1562 (2001)
Nandedkar, A.V., Biswas, P.K.: A fuzzy min-max neural network classifier with compensatory neuron architecture. IEEE Trans. Neural Netw. 18(1), 42–54 (2007)
Kondekar, M.H., Kulkarni, U.V., Chowhan, S.S.: Finger print recognition using extended fuzzy hypersphere neural network. J. Comput. 3(4), 101–105 (2011)
Chowhan S.S., Kulkarni U.V., Shinde G.N.: Iris recognition using modified fuzzy hypersphere neural network with different distance measures. Int. J. Adv. Comput. Sci. Appl. 2(6), 130–134 (2011)
Sonar D.N., Kulkarni U.V.: Pruned fuzzy hypersphere neural network for lung cancer classification. Int. J. Comput. Appl. 157, 36–39 (2017)
Hu, J., Luo, Y.: Integration of learning algorithm on fuzzy min-max neural networks. J. Shanghai Jiaotong Univ. (Sci.) 22, 733–741 (2017)
Mane, D.T., Kulkarni, U.V.: Modified fuzzy hypersphere neural network for pattern classification using supervised clustering. Procedia Comput. Sci. 143, 295–302 (2018)
Kulkarni, A.B., Bonde, S.V., Kulkarni, U.V.: Class-specific fuzzy hypersphere neural network. Procedia Comput. Sci. 143, 285–294 (2018)
Waghmare, J.M., Kulkarni U.V.: Unbounded recurrent fuzzy min-max neural network for pattern classification. In: International Joint Conference on Neural Networks (IJCNN), Budapest, Hungary (2019)
Kumar, A., Sai Prasad, P.S.V.S.: Hybridization of fuzzy min-max neural networks with kNN for enhanced pattern classification. In: Singh, M., Gupta, P.K., Tyagi, V., Flusser, J., Ören, T., Kashyap, R. (eds.) ICACDS 2019. CCIS, vol. 1045, pp. 32–44. Springer, Singapore (2019). https://doi.org/10.1007/978-981-13-9939-8_4
Tran, T.N., Vu, D.M., Tran, M.T., et al.: The Combination of fuzzy min-max neural network and semi-supervised learning in solving liver disease diagnosis support problem. Arab. J. Sci. Eng. 44, 2933–2944 (2019)
Kulkarni, A., Kulkarni, N.: Fuzzy neural network for pattern classification. Proc. Comput. Sci. 167, 2606–2616 (2020)
Mane, D.T., Kshirsagar, J.P., Kulkarni, U.V.: Modified quick fuzzy hypersphere neural network for pattern classification using supervised clustering. In: Reddy, V.S., Prasad, V.K., Wang, J., Reddy, K.T.V. (eds.) ICSCSP 2019. AISC, vol. 1118, pp. 227–235. Springer, Singapore (2020). https://doi.org/10.1007/978-981-15-2475-2_21
Mahindrakar, M.S., Kulkarni, U.V.: Unbounded Fuzzy Hypersphere Neural Network Classifier. Manuscript Submitted for Publication
Dua, D.: UCI machine learning repository (2018). http://archive.ics.uci.edu/ml
Davtalab, R., Dezfoulian, M.H., Mansoorizadeh, M.: Multi-level fuzzy min-max neural network classifier. IEEE Trans. Neural Netw. Learn. Syst. 25(3), 470–482 (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2022 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Mahindrakar, M.S., Kulkarni, U.V. (2022). Optimized Fuzzy Hypersphere Neural Network with Online Adaptation Capability. In: Iyer, B., Crick, T., Peng, SL. (eds) Applied Computational Technologies. ICCET 2022. Smart Innovation, Systems and Technologies, vol 303. Springer, Singapore. https://doi.org/10.1007/978-981-19-2719-5_7
Download citation
DOI: https://doi.org/10.1007/978-981-19-2719-5_7
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-19-2718-8
Online ISBN: 978-981-19-2719-5
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)