Machine learning such as self-organizing feature maps (SOMs) is a commonly used technique for clustering and data dimensionality reduction. In fact, their inherent property of topology preservation and unsupervised learning of processed data put them in the front of candidates for data reduction. However, the high computational cost of SOMs limits their use to offline approaches and makes the on-line real-time high-performance SOM processing more challenging. This chapter focus on the large-scale distributed and scalable SOM model adapted for distributed computing nodes and present the main challenges for its adoption in the resources limited environments.

Chapter Contents:

• 16.1 Introduction
• 16.2 Related work
• 16.3 Background
• 16.4 Proposed SOM model
• 16.4.1 NoC router model
• 16.4.2 Neuron model
• 16.4.3 CWS model
• 16.4.4 NI model
• 16.5 Results and discussion
• 16.6 Conclusion
• References

Inspec keywords: pattern clustering; learning (artificial intelligence); data reduction; self-organising feature maps; distributed processing

Other keywords: distributed computing nodes; large-scale distributed scalable SOM-based architecture; high-dimensional data reduction; data clustering; topology preservation; unsupervised learning; data dimensionality reduction; self-organizing feature maps

Subjects: Data handling techniques; Knowledge engineering techniques; Neural computing techniques; Distributed systems software; General and management topics