High-dimensional biological data, statistical

The SOM is a model that emphasizes particular aspects of biological neural nets, especially those ordered maps found experimentally in the cortex. The SOM captures some of the fundamental processing principles of the brain. The SOM is a model of unsupervised machine learning and also an adaptive knowledge representation scheme. The SOM is a tool for statistical analysis and visualization. The SOM is both a projection method that maps high-dimensional data space into low-dimensional space, and a clustering method in which similar data samples tend to be mapped to nearby neurons. 

---