This was the beginning of our foray into the area of unsupervised learning. Unsupervised learning algorithms are the category of Machine Learning algorithms where the data we’ve been handed are not labeled, so we no longer are at the vantage point of deciphering what the intent of a given data set is to begin with.
Unsupervised learning problems are typically dealt by grouping data into categories. Earlier this week, we worked through the well-known clustering algorithm called the K-means clustering. During the second part of the week, we explored the realm of dimensionality reduction where we take a reductionist approach to data for the sake of ease of human comprehension and to speed up the computational time of learning algorithms.