Performance Improvement of Extension Neural Network Using Data Selection Method Based on Shadowed Sets
-
摘要: 为了改进可拓神经网络的性能,提出一种基于阴影集的数据选择方法.通过该方法获取用于训练可拓神经网络的训练样本,进而改进可拓神经网络的性能.针对可拓神经网络的特点,选择核数据和边界数据作为可拓神经网络的训练样本;利用基于阴影集的数据选择方法,可以自动获取核数据和边界数据.实验结果表明,与传统可拓神经网络相比,改进的可拓神经网络不仅节约了训练时间,而且网络的泛化能力和分类识别准确度得到了有效提高.Abstract: To improve the performance of extension neural network(ENN),a data selection method based on shadowed sets was proposed.This method was used to obtain training sample data for improving the performance of ENN.According to the characteristics of ENN,core data and boundary data were selected as training data for ENN;using shadowed-sets-based data selection method,core data and boundary data could be captured automatically.Experimental results indicate that the learning speed of the improved extension neural network(IENN) is faster than traditional ENN.Moreover,the generalization ability and the recognition accuracy are improved effectively.
-
-
[1] 魏海坤,徐嗣鑫,宋文忠.神经网络的泛化理论和泛化方法[J].自动化学报,2001,27(6):806-815.WEI Hai-kun,XU Si-xin,SONG Wen-zhong.Generalization theory and generalization methods for neuralnetworks[J].Acta Automatica Sinica,2001,27(6):806-815.(in Chinese) [2] ZHOU Yu,WU Ya-li.Analyses on influence of trainingdata set to neural network supervised learning performance[J].Advances in Computer Science,Intelligent Systemand Environment,2011,106:19-25.
[3] GUAN D,YUAN W,LEE Y K,et al.Improvingsupervised learning performance by using fuzzy clusteringmethod to select training data[J].Journal of Intelligent&Fuzzy Systems,2008,19:321-334.
[4] 马翔,陈新楚,王邵伯.均匀设计法在RBF神经网络样本优选中的应用[J].模式识别与人工智能,2005,18(2):252-255.MA Xiang,CHEN Xin-chu,WANG Shao-bo.Applicationof the uniform design to the optimal selection of samples forRBF neural network[J].Pattern Recognition and ArtificialIntelligence,2005,18(2):252-255.(in Chinese) [5] PEREIRA C E.Learning vector quantization with trainingdata selection[J].IEEE Trans on Pattern Analysis andMachine Intelligence,2006,28(1):157-162.
[6] FUKUMIZU F.Statistical active learning in multilayerperceptrons[J].IEEE Transactions on Neural Networks,2000,11(1):17-26.
[7] PEDRYCZ W.Interpretation of cluster in the framework ofshadowed sets[J].Pattern Recognition Letters,2005,26(15):2439-2449.
[8] PEDRYCZ W.From fuzzy sets to shadowed sets:interpretation and computing[J].International Journal ofIntelligence System,2009,24(1):48-61.
[9] WANG M H,HUNG C P.Extension neural network andits applications[J].Neural Networks,2003,16(5):779-784.
[10] 周玉,钱旭,张俊彩.可拓神经网络研究综述[J].计算机应用研究,2010,27(1):1-5.ZHOU Yu,QIAN Xu,ZHANG Jun-cai.Survey andresearch of extension neural network[J].ApplicationResearch of Computers,2010,27(1):1-5.(inChinese) [11] CAI Wen.Extension theory and its applications[J].Chinese Science Bulletin,1999,44(17):1538-1548.
[12] 杨春燕,蔡文.可拓工程[M].北京:科学出版社,2007:18-97. [13] ZHOU Yu,PEDRYCZ W,QIAN Xu.Application ofextension neural network to safety status patternrecognition of coal mines[J].Journal of Central SouthUniversity of Technology,2011,18(3):633-641.
[14] WANG W H.Partial discharge pattern recognition ofcurrent transformers using an ENN[J].IEEE Trans onPower Delivery,2005,20(3):1984-1990.
[15] LAI Y H,CHE H C.Modeling patent legal value byextension neural network[J].Expert Systems WithApplications,2009,36(7):10520-10528.
[16] WANG M H.Extension neural network-type 2 and itsapplications[J].IEEE Trans on Neural Networks,2005,16(6):1352-1361.
[17] CATTENEO G,CIUCCI D.An algebraic approach toshadowed sets[J].Electronic Notes in TheoreticalComputer Science,2003,82(4):64-75.
[18] MITRA S,PEDRYCZ W,BARMAN B.Shadowed C-means:integrating fuzzy and rough clustering[J].Pattern Recognition,2010,43(4):1282-1291.
[19] BARMAN B,MITRA S,PEDRYCZ W.Shadowedclustering for speech data and medical image segmentation[J].Lecture Notes in Computer Science,2008,5306:475-484.
[20] GRKAN E,ERKMEN.I,ERKMEN A M.Two-wayfuzzy adaptive identification and control of a flexible-jointrobot arm[J].Information Sciences,2002,145(1/2):13-43.
计量
- 文章访问数: 1
- HTML全文浏览量: 0
- PDF下载量: 0
下载:
