Machine Learning: Clustering Association Rules

Name: Machine Learning: Clustering Association Rules
Author: Briana Brownell

Navigating association rules and their applications in retail, e-commerce, and media platforms.

Preview the first lesson free — get full access to all 4 lessons.

Course: On-Demand

In this chapter, we're gonna talk about association rules, and the kinds of problems that this kind of modeling can solve. A way to think about association rules is you want to figure out associations between different variables, and which ones can predict the other. So the most common problem that association rule learning is trying to solve is what's called market basket analysis. So what we want to figure out is if a shopper purchases a certain item, what other items are they most likely to be purchasing at the same time? So here, I've got an example of five baskets from five shoppers, and we've got a bunch of different items that each shopper has purchased. You can imagine this data set coming from transactions at a grocery store. You could use it in an eCommerce website. You could even use it inside of a technology system where you're looking at the features that somebody used rather than the items that they purchased. In a media site, you could think of it as the TV shows or movies that that person has watched. But for the example, we're gonna look at basket analysis in terms of purchases, and things that are purchased together. So just by looking at this, it's difficult to see what the relationships might be between the variables. And especially when you have a large data set, being able to tease out those relationships can be really, really challenging. And so association rules can help us do that. And there are a number of things that you could use association rules for. So one is recommendations. So if you notice that there are items that are frequently purchased together on an online store, you may want to recommend that second item to maybe prompt that person to consider purchasing that item or make it easier for them to find it. So if this person already wants to purchase that second item, having it right there is gonna make their experience better. You can definitely see this in the grocery store where the stores starting to group items together that are associated. And I am willing to bet that the analysis that they used uses association rules. Another way that you can use this is to optimize promotions. For example, if you notice that there are two items that are very strongly linked, and you put item A on sale, then you can expect that item B will also have a number of sales if it's bought together. So you could think of it as items that are related in a recipe, let's say. So if people see item A on sale, and that makes them think of making a specific recipe, then it's pretty easy to see the reason that item B would be closely related to that one as well. So by using association rules, you can encourage shoppers to purchase groups of things in your store, and hopefully give them a deal on something that they want to make. So now you should have an idea of what association rules are, and the reasons you might want to do this analysis, as well as strategies that you can take to use this kind of analysis in practice.

Intermediate Provider : Briana Brownell 4 Lessons · 16m in Arabic, German, English, Spanish, French, Portuguese, Urdu, Chinese

Course Description

In these lessons, you’ll learn about association rules and how they can apply to retail stores, e-commerce websites and media platforms. In turn, we’ll explore the three most important measures--support, confidence, and lift--as well as how to interpret each.

This course will also be a guide for how to navigate and apply association rules, particularly in setting up data for association rules to facilitate and maximize their interpretation and use. Finally, you will learn how to apply association rules to business strategy.

What You'll Learn

Apply association rules to retail stores, e-commerce websites, and media platforms
Interpret the three key measures of support, confidence, and lift
Set up data correctly to facilitate and maximize association rule interpretation
Navigate and apply association rules in practice
Apply association rules to business strategy

Key Takeaways

Association rules can be applied to retail stores, e-commerce websites, and media platforms.
Support, confidence, and lift are the three most important measures used to evaluate association rules.
Setting up data appropriately helps facilitate and maximize the interpretation and use of association rules.
Association rules can be applied to inform business strategy.

Frequently Asked Questions

What will I learn in this course?

You will learn how to apply association rules, interpret them, and set up data for association rules, including the three key measures of support, confidence, and lift.

What topics does this course cover?

The course covers association rules, typical data setup, support, confidence and lift, and interpretation.

Where can association rules be applied?

Association rules can apply to retail stores, e-commerce websites, and media platforms, and can be applied to business strategy.

What skills does this course help develop?

It supports skills in association rule learning, cluster analysis, data mining, data mining methods, machine learning algorithms, and relational data mining.