Chapter 11Discovering Patterns with Association Rules
In Chapters 4 through 8, we introduced several supervised machine learning approaches. With those approaches, we used previously labeled data to train a model that we then used to assign labels to unlabeled data. In Chapters 9 and 10, we discussed several of the common approaches used in evaluating and improving the performance of a supervised learning. In the next two chapters, we will introduce two unsupervised learning techniques. Unsupervised learning differs from supervised learning in that with unsupervised learning, there are no previously labeled examples to learn from. With unsupervised learning, we are not attempting to make a prediction; instead, we are looking for new and interesting patterns and insights in the data.
In this chapter, we introduce the first of the two unsupervised machine learning techniques we cover in this book—association rules. Association rules are often used to discover patterns that exist within a set of transactions. These transactions can be retail transactions that occur at a point of sale, they can be symptoms that are observed when certain medications are administered to patients during a drug trial, or they can be any set of items or events that occur together at distinct points in time.
By the end of this chapter, you will have learned the following:
- The basic ideas behind the association rules approach
- The different ways to evaluate and quantify the strength of association rules ...
Get Practical Machine Learning in R now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.