Introduction

Most data science practitioners would agree that natural language processing, including topic modeling, is toward the cutting edge of data science and is an active research area. We now understand that topic models can, and should, be leveraged wherever text data could potentially drive insights or growth, including in social media analyses, recommendation engines, and news filtering. The preceding chapter featured an exploration of the fundamental features of topic models and two of the major algorithms. In this chapter, we are going to change direction entirely.

This chapter takes us into the retail space to explore a foundational and reliable algorithm for analyzing transaction data. While this algorithm might not be on the cutting edge or in the catalog of the most popular machine learning algorithms, it is ubiquitous and undeniably impactful in the retail space. The insights it drives are easily interpretable, immediately actionable, and instructive for determining...