A self-organizing map (SOM) or Kohonen network or Kohonen map is a type of artificial neural network that is trained using unsupervised learning to produce a low-dimensional (typically two-dimensional), discretized representation of the input space of the training samples, called a map, which preserves the topological properties of the input space (

Wikipedia).

SOM is useful for the dimensionality reduction, data visualization and cluster analysis. In this course material, we outline the mechanisms underlying the approach. We focus on its practical aspects (e.g. various visualization possibilities, prediction on a new instance, extension of SOM to the clustering task,…).

Illustrative examples in

**R** (kohonen package) and

**Tanagra** are briefly presented.

**Keywords:** som, self organizing map, kohonen network, data visualization, dimensionality reduction, cluster analysis, clustering, hierarchical agglomerative clustering, hac, two-step clustering, R software, kohonen package

**Components**: KOHONEN-SOM

**Slides**:

Kohonen SOM
**References**:

Wikipedia, "

Self-organizing map".