SOM Algorithm Each data from data set recognizes themselves by competeting for representation. SOM mapping steps starts from initializing the weight vectors. The Self-Organizing Map (SOM) has applications like dimension reduction, data clustering, image analysis, and many others. In conventional SOM, the weights. A self-organizing map (SOM) is a clustering technique that helps you the algorithm has converged, we can plot the evolution of the SOM's.


Author: Richard Jacobi
Country: Vanuatu
Language: English
Genre: Education
Published: 6 November 2016
Pages: 201
PDF File Size: 18.35 Mb
ePub File Size: 40.83 Mb
ISBN: 275-3-37479-483-6
Downloads: 94986
Price: Free
Uploader: Richard Jacobi



This is the essence of the SOM algorithm. The neurons represent the inputs with reference vectorsthe components of which correspond to synaptic weights.

Iterative process of an SOM.


The learning rate and BMU radius should be tuned via validation. If self organizing map algorithm for both are too high, neurons will be shoved around constantly without settling down.

But if values are too low, the analysis will take too long as neurons inch towards their optimal positions.

  • The self-organizing map algorithm
  • Self-organizing map - Wikipedia
  • There was a problem providing the content you requested
  • Self-organizing map
  • Bibliography

Hence, it is ideal to start with larger learning rate and BMU radius first, before reducing them over time. Another feature that we need to validate is the optimal number of neurons in the grid. Recall that self organizing map algorithm each neuron has multiple data points associated with it, it can be treated as a mini-cluster.

We can thus validate each neuron to self organizing map algorithm if its associated data points correspond to known sub-clusters of, say, consumer profiles.

Self-Organizing Maps Tutorial – Algobeans

In order for such clusters to be distilled however, there should be fewer neurons than data points, so that similar data points can be mapped to each neuron.

One thing to note before we apply SOM: Variables measured in different units could interfere with the speed and accuracy of our analysis. For example, a variable measured in centimeters would have a value times self organizing map algorithm than the same one measured in meters.


To prevent any variable from overpowering the others, we need to standardize all self organizing map algorithm. Standardization is analogous to expressing each variable in terms of percentiles, meaning to shift them onto a uniform standard scale, so that they are of the same measurement unit.

Limitations of SOM SOM simplifies datasets with many variables, which is useful for visualization and identifying clusters.

Self-Organizing Map - Clever Algorithms: Nature-Inspired Programming Recipes

The system is initialized to vectors within the domain although is only exposed to samples within the pre-defined shape during training.

The expectation is that the system will model the shape based on the observed samples.

The algorithm is an implementation of the basic Self-Organizing Map algorithm based on the description in Chapter 3 of the seminal book on the technique [ Kohonen ].

Error rates are presented to the console, and the codebook vectors themselves are described before and after training. More neurons point to regions with high training sample concentration and fewer where the samples are scarce.

Originally, SOM was not formulated self organizing map algorithm a solution to an optimisation problem. Nevertheless, there have been several attempts self organizing map algorithm modify the definition of SOM and to formulate an optimisation problem which gives similar results.

In the sense that a GTM explicitly requires a smooth and continuous mapping from the input space to the map space, it is self organizing map algorithm preserving. However, in a practical sense, this measure of topological preservation is lacking.

Related Post