Product Analytics: Applied Data Science Techniques for Actionable Consumer Insights

7. Modeling the User Space: k-Means and PCA

In the last section, we covered distributions, metrics, and A/B testing. These techniques are important for us to operationalize concepts. In this chapter, we’ll be learning tools to group similar variables, which can be very useful when we want to create metrics to represent a broad conceptual idea. Clustering can also be used to narrow down a large set of variables or group similar observations together.

In this section, we will expand our toolkit to include clustering and predictive algorithms. The first two chapters will focus on machine learning tools, and the last chapter will discuss demographic prediction (projection) techniques. This chapter is intended for business practitioners new to clustering ...