Title :
A growing self-organizing algorithm for dynamic clustering
Author :
Ohta, Ryuji ; Saito, Toshimichi
Author_Institution :
Dept. of Electron. & Electr. Eng., Hosei Univ., Japan
Abstract :
Presents a growing self-organizing algorithm for dynamic clustering. Controlling a signal counter of each cell, the network can grow. Also, if there exists undesired cell for the clustering, the cell can be deleted virtually. Our algorithm can reinforce the clustering function and the learned network can adapt flexibly to time-variant input space
Keywords :
pattern clustering; self-organising feature maps; unsupervised learning; clustering function; dynamic clustering; growing self-organizing algorithm; learned network; signal counter; time-variant input space; Clustering algorithms; Counting circuits; Data mining; Force control; Heuristic algorithms; Image coding; Neural networks; Organizing; Speech recognition; Unsupervised learning;
Conference_Titel :
Neural Networks, 2001. Proceedings. IJCNN '01. International Joint Conference on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-7044-9
DOI :
10.1109/IJCNN.2001.939065
