Scribd
Upload
7 views15 pages

Unit4. SOM

The Self-Organizing Map (SOM) is an unsupervised learning algorithm designed for dimensionality reduction and data visualization, developed by Teuvo Kohonen in 1982. It effectively maps high-dimensional data into lower-dimensional spaces while preserving the topological structure, making it useful for clustering and pattern recognition. Key features include topology preservation, dimensionality reduction, and the ability to learn without labeled data.

Uploaded by

Deepanshu Tyagi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
7 views15 pages

Unit4. SOM

The Self-Organizing Map (SOM) is an unsupervised learning algorithm designed for dimensionality reduction and data visualization, developed by Teuvo Kohonen in 1982. It effectively maps high-dimensional data into lower-dimensional spaces while preserving the topological structure, making it useful for clustering and pattern recognition. Key features include topology preservation, dimensionality reduction, and the ability to learn without labeled data.

Uploaded by

Deepanshu Tyagi
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 15

Self Organising Map (SOM)

Unit 4
Dr. Sushil Kumar
Associate Professor
Self-Organizing Map (SOM)
• A SOM is an unsupervised learning algorithm used for dimensionality reduction
and data visualization.
• Developed by Teuvo Kohonen in 1982.
• SOMs are particularly effective in mapping high-dimensional data into lower-
dimensional spaces (typically 2D) while preserving the topological structure of
the input data.
• SOMs useful for clustering and pattern recognition tasks.
Key Features
1. Topology Preservation: Similar data points in the input space remain close to
each other in the map.
2. Dimensionality Reduction: It reduces complex data into a simpler, visual
representation.
3. Unsupervised Learning: SOMs learn without labeled data, discovering hidden
patterns or structures.
Structure
Steps used in SOM
1. Initialization
2. Input Vector
3. Best Matching Unit (BMU)
4. Weight Update Rule
5. Training Process
Initialization
Input Vector
Best Matching Unit (BMU)
Weight Update Rule
Training Process
Example:
Neighborhood Function
Weight Update Rule
Decay of Parameters

You might also like