Title :
The use of formal measures for the training of hierarchical Kohonen maps
Author :
Weierich, Peter ; Von Rosenberg, Michael
Author_Institution :
Knowledge Process. Res. Group, Bavarian Res. Center for Knowledge-Based Syst., Erlangen, Germany
fDate :
27 Jun-2 Jul 1994
Abstract :
We present a means of evaluating the relative recognition rate of unsupervised classifiers. Self organizing maps (so called Kohonen Maps) have some important features. They can classify data, reduce the dimension of input vectors and can be trained using unlabeled data. In our efforts to develop an automatic, context dependent classifier we highly relied on these features. Using unlabeled time series data, we had no way to compare different topologies with respect to “correct classifications”. We solved this problem by introducing a formal measure we called pseudo classes. This is an elegant, but heuristic method which can also be applied on other fields with context-dependent data
Keywords :
self-organising feature maps; time series; unsupervised learning; context dependent classifier; context-dependent data; data classification; heuristic method; hierarchical Kohonen maps; input vectors; neural network training; pseudo classes; recognition rate; self organizing maps; unlabeled data; unlabeled time series data; unsupervised classifiers; Artificial neural networks; Automatic testing; Biological neural networks; Delay effects; Fault detection; Humans; Knowledge based systems; Self organizing feature maps; Signal detection; Topology;
Conference_Titel :
Neural Networks, 1994. IEEE World Congress on Computational Intelligence., 1994 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-1901-X
DOI :
10.1109/ICNN.1994.374245
