In the previous chapters, I stressed the importance of understanding the mathematics behind algorithms. Here's a recap. We started with linear regression, followed by a Naïve Bayes classifier. Then, the topics dovetailed into one of the more complex topics in data science: time series. We then detoured and discussed clustering by means of K-means. This was followed by two chapters on neural networks. In all these chapters, I explained the mathematics behind these algorithms, and showed that, with much surprise, the programs yielded are short and simple.

The purpose of this book is to walk a delicate line between the math and the implementations. I hope I have provided enough information so that you have an understanding of the mathematics and how they may be useful. The projects are real projects, but often they are in various forms, simplified and rather academic. And so, it may be a bit of a surprise that this chapter will not contain many mathematical explanations. Instead, this chapter is aimed at guiding readers through more real-world scenarios.

In the previous chapter, we discussed facial detection. Given an image, we want to find the faces. But who are they? In order to know who the faces belong to, we'd need to perform facial recognition.