Now, let's start by talking about what we're doing with dimensionality reduction. In this section, I will introduce dimensionality reduction techniques as an unsupervised learning method, and discuss what the objectives of dimensionality reduction are and what it is used for.
Dimensionality reduction is considered an unsupervised learning method, since there is no target variable that we are trying to predict. The other unsupervised learning method we looked at was clustering, which was unsupervised learning's analogy to classification.
Dimensionality reduction is the unsupervised learning analogy to regression. We are trying to discover features, often in Euclidean space, that we do not directly observe in data, yet we believe influence patterns seen in it. Perhaps you may recall the curse of dimensionality. This is a phenomenon...