An OrdinationFuzzy MinMax Neural Network Classifier on Unlabelled Pattern Classification
HU Jing, YANG Jing, GAO Jun
Laboratory of Image Information Processing, School of Computer and Information, Hefei University of Technology, Hefei 230009 Center for Biomimetic Sensing and Control Research, Institute of Intelligent Machines, Chinese Academy of Sciences, Hefei 230031
Abstract An ordinationfuzzy minmax neural network(OFMM) based on nonmetric multidimensional scaling (MDS) is proposed to solve the classification problems of unlabelled input pattern. Firstly, all the input patterns are sorted by MDS to get their similarity measures. Then these measures are used to supervise the following expansion and contraction stage of hyperboxes for classification. OFMM shows the improvements in the validity of unlabelled patterns classification, the network structure, and training time. The experimental results on standard dataset demonstrate that OFMM is a practical and effective classifier which is superior to the traditional generalfuzzy minmax neural network (GFMM).
[1] Shahshahani B M, Landgrebe D A. The Effect of Unlabeled Samples in Reducing the Small Sample Size Problem and Mitigating the Hughes Phenomenon. IEEE Trans on Geoscience and Remote Sensing, 1994, 32 (5): 10871095 [2] Anguita D, Ridella S, Rivieccio F, et al. Unsupervised Clustering and the Capacity of Support Vector Machines // Proc of the IEEE International Joint Conference on Neural Networks. Budapest, Hungary, 2004, Ⅲ: 20232028 [3] Kothari R, Jain V. Learning from Labeled and Unlabeled Data Using Minimal Number of Queries. IEEE Trans on Neural Networks, 2003, 14 (6): 14961505 [4] Fardanessh M T, Ersoy O K. Classification Accuracy Improvement of Neural Network Classifiers by Using Unlabeled Data. IEEE Trans on Geoscience and Remote Sensing, 1998, 36(3): 10201025 [5] Simpson P K. Fuzzy MinMax Neural Networks-Part 1: Classification. IEEE Trans on Neural Networks, 1992, 3(5): 766786 [6] Shepard R N. The Analysis of Proximities: The Multidimensional Scaling with an Unknown Distance Function. Psychometrika, 1962, 27(5): 125140 [7] Kryskal J B. Nonmetric Multidimensional Scaling: A Numerical Method. Psychometrika, 1964, 29(3): 115129 [8] Gabrys B, Bargiela A. General Fuzzy MinMax Neural Networks for Clustering and Classification. IEEE Trans on Neural Networks, 2000, 11(3): 769783 [9] Yu Shixiao. The Application of NonMetric Multidimensional Scaling in Community Classification. Journal of Plant Ecology, 1995, 19(2): 128136(in Chinese) (余世孝.非度量多维测度及其在群落分类中的应用.植物生态学报, 1995, 19(2): 128136) [10] Blake C, Keogh E, Merz C. UCI Repository of Machine Learning Databases[DB/OL]. [19981201]. http://www.ics.uci.edu/~melearn/ML Repository.html [11] Gao Ju. The Principle of Neural Networks and Its Simulated Experiments. Beijing, China: China Machine Press, 2003(in Chinese) (高 隽.人工神经网络原理及仿真实例.北京:机械工业出版社, 2003) [12] Xiao Li, Jin Yuanping, Xu Hongbin, et al. A Multidimensional Scaling Based Algorithm for Fast Mining Association Rules. Journal of Software, 1999, 10(7): 749753(in Chinese) (肖 利,金远平,徐宏炳,等.基于多维标度的快速挖掘关联规则算法.软件学报, 1999, 10(7): 749753) [13] Levine S H, Kreifeldt J G, Chuang M C. Point Pattern Representation Using Imprecise, Incomplete, Nonmetric Information. IEEE Trans on Systems, Man and Cybernetics, 1994, 24(2): 222233 [14] Ross G, Chalmers M. A Visual Workspace for Hybrid Multidimensional Scaling Algorithms // Proc of the IEEE Symposium on Information Visualization. Seattle, USA, 2003: 9196
2007, Vol. 20 
