The feature map represented by the set of weight vectors of the basic SOM (Self-Organizing Map) provides a good approximation to the input space from which the sample vectors come. But the timedecreasing learning rate and neighborhood function of the basic SOM algorithm reduce its capability to adapt weights for a varied environment. In dealing with non-stationary input distributions and changi...

In this paper, we propose a framework based on a subsymbolic approach for the support of exploratory activities in a hyperspace. By using it, it is possible to express the semantic features of the whole hyperspace and the states of exploratory activities in topological order. This approach is applied to generate the navigation information for the exploratory activity. The space explored is chan...

The technique of searching in content-based image retrieval has been actively studied in recent years. However, this technique cannot give user an overview of the database. In this paper, we propose a browsing technique using Kohonen’s Self-Organizing Map to retrieve general color image database effectively. Both chromatic and textural feature of images are analyzed to represent the content of ...

This paper introduces a hierarchical Self-Organizing Feature Map (SOFM). The partial maps consist of individual numbers of neurons, which makes a cluster analysis with di erent degrees of resolution possible. A de nition of a special Mahalanobis space of the data set during the learning improves the properties concerning clusters with low density.

So far the bottleneck of Chinese character recognition, especially handwritten character recognition, still lies in the effectiveness of featureextraction to cater for various distortions and position shifting. In ths paper, a novel method is proposed by applying a set of Gabor spatial filters with different directions and spatial frequencies to character images, in an effort to reach the optim...

So far the bottleneck of Chinese character recognition, especially handwritten character recognition, still lies in the effectiveness of featureextraction to cater for various distortions and position shifting. In ths paper, a novel method is proposed by applying a set of Gabor spatial filters with different directions and spatial frequencies to character images, in an effort to reach the optim...