Community SOM (CSOM): An Improved Self-Organizing Map Learning Technique

In self-organizing map, the neurons in the 1-neighborhood of winner neuron are called community of winner neurons. The neuron which turns out to be the winner for the least number of times after a specified number of iterations has been named here as the weakest neuron. The neurons which are either the weakest or the farthest in the community of winner neurons are not getting enough exposure, which decreases their learning efficiency a lot. A community self-organizing map has been proposed here, which facilitates the learning of the weakest and farthest neuron in the community of winner neurons using a different process, thereby increasing the overall learning.

[1]  David Zipser,et al.  Feature Discovery by Competive Learning , 1986, Cogn. Sci..

[2]  J. Rubner,et al.  A Self-Organizing Network for Principal-Component Analysis , 1989 .

[3]  John W. Sammon,et al.  A Nonlinear Mapping for Data Structure Analysis , 1969, IEEE Transactions on Computers.

[4]  Esa Alhoniemi,et al.  Clustering of the self-organizing map , 2000, IEEE Trans. Neural Networks Learn. Syst..

[5]  Yi Lu Murphey,et al.  Multi-class pattern classification using neural networks , 2007, Pattern Recognit..

[6]  R. Tibshirani,et al.  Adaptive Principal Surfaces , 1994 .

[7]  Michael E. Tipping,et al.  Probabilistic Principal Component Analysis , 1999 .

[8]  Nikhil R. Pal,et al.  Two efficient connectionist schemes for structure preserving dimensionality reduction , 1998, IEEE Trans. Neural Networks.

[9]  Erkki Oja,et al.  Rival penalized competitive learning for clustering analysis, RBF net, and curve detection , 1993, IEEE Trans. Neural Networks.

[10]  Antanas Verikas,et al.  Feature selection with neural networks , 2002, Pattern Recognit. Lett..

[11]  Yoshifumi Nishio,et al.  Shooting SOM and its Application for Clustering , 2006 .

[12]  Yiu-ming Cheung,et al.  On Rival Penalization Controlled Competitive Learning for Clustering with Automatic Cluster Number Selection , 2005, IEEE Trans. Knowl. Data Eng..

[13]  Teuvo Kohonen,et al.  Self-Organizing Maps , 2010 .

[14]  Yu-ming Cheung.  Rival penalization controlled competitive learning for data clustering with unknown cluster number , 2002, Proceedings of the 9th International Conference on Neural Information Processing, 2002. ICONIP '02..

[15]  Teuvo Kohonen,et al.  Self-organized formation of topologically correct feature maps , 2004, Biological Cybernetics.

[16]  Duane DeSieno,et al.  Adding a conscience to competitive learning , 1988, IEEE 1988 International Conference on Neural Networks.

[17]  T. Kohonen.  Self-organized formation of topographically correct feature maps , 1982 .

[18]  Min Han,et al.  Efficient clustering of radial basis perceptron neural network for pattern recognition , 2004, Pattern Recognit..

[19]  Yoshifumi Nishio,et al.  Behavior of Fatigable SOM and its Application to Clustering , 2006, The 2006 IEEE International Joint Conference on Neural Network Proceedings.

[20]  Nithiroth Pornsuwancharoen,et al.  A new technique Gray scale display of input data using shooting SOM and genetic algorithm , 2012 .

[21]  Yiu-ming Cheung,et al.  Rival-Model Penalized Self-Organizing Map , 2007, IEEE Transactions on Neural Networks.