15 Dimension reduction with networks and local structure: self-organizing maps and locally-linear embedding
This chapter covers:
- How to create self-organizing maps to reduce dimensionality
- Creating locally-linear embeddings of high-dimensional data
In this chapter, we’re continuing with dimension reduction: the class of machine learning task focused on representing the information contained in a large number of variables, in a smaller number of variables. As we learned in the last two chapters, there are multiple possible ways for us to reduce the dimensieqons of a dataset. Which dimension reduction algorithm works best for you depends on the structure of your data, and what you’re trying to achieve. Therefore, in this chapter I’m going to add two more non-linear dimension reduction algorithms to your ever-growing machine learning toolbox:
- self-organizing maps (SOMs)
- locally-linear embedding (LLE)
Both the SOM and LLE algorithms reduce a large dataset into a smaller, more manageable number of variables, but work in very different ways. The SOM algorithm creates a two-dimensional grid of nodes, like grid-references on a map. Each case in the data is placed into a node, and then shuffled around the nodes to put cases that are more similar to each other in the original data, close together on the map.