Brief review of self-organizing maps

As a particular type of artificial neural networks, self-organizing maps (SOMs) are trained using an unsupervised, competitive learning to produce a low-dimensional, discretized representation of the input space of the training samples, called a feature map. Such a map retains principle features of the input data. Self-organizing maps are known for its clustering, visualization and classification capabilities. In this brief review paper basic tenets, including motivation, architecture, math description and applications are reviewed.