Applications of the Growing Self Organizing Map on high dimensional data

The Growing Self Organizing Map (GSOM) is a dynamic variant of the Self Organizing Map (SOM). It has been mainly used on low dimensional data sets. In this paper the GSOM is applied on high dimensional data sets and its performance is evaluated. Several modifications to the original GSOM algorithm are presented that enable the GSOM to be applied on high dimensional data .The modified version of the GSOM is called the High Dimensional Growing Self Organizing Map (HDGSOM).

[1]  Damminda Alahakoon,et al.  Controlling the spread of dynamic self-organising maps , 2004, Neural Computing & Applications.

[2]  Andreas Rauber,et al.  The growing hierarchical self-organizing map , 2000, Proceedings of the IEEE-INNS-ENNS International Joint Conference on Neural Networks. IJCNN 2000. Neural Computing: New Challenges and Perspectives for the New Millennium.

[3]  Saman K. Halgamuge,et al.  A self-growing cluster development approach to data mining , 1998, SMC'98 Conference Proceedings. 1998 IEEE International Conference on Systems, Man, and Cybernetics (Cat. No.98CH36218).

[4]  T. Kohonen.  Analysis of a simple self-organizing process , 1982, Biological Cybernetics.

[5]  Bala Srinivasan,et al.  Visualising Cluster Separation with Dynamic SOM Tree , 2000 .

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