Skip To Content

What is Association Mining (and how can your business use it)?

Reading time: 3 minutes
Association Mining

By Earl Sires

Every business collects data: sales history, pricing trends, employee satisfaction, and more. Within that data lies insight: valuable information you can apply to improve your business. Often, that insight isn’t immediately apparent; instead, it requires a deeper investigation of the patterns and trends within your data. That process of discovery is known as data mining.

While there are many methods of mining insight from data, one of the most ‘fruitful’ (the pun will make more sense in a moment, I promise) is called association mining. 

What is Association Mining?

“Association mining,” sometimes known as market basket analysis, describes the process of finding correlative links and patterns within variables in a set of data.

Here’s a straightforward example. Imagine you have a dataset containing the purchases of all the customers who shop at your grocery store. You know what groups of items each customer purchased in the past. Through association mining, you can make larger inferences about what groups of items tend to be purchased together. This information can guide your strategic business decisions. 

Sometimes the links are obvious. If someone buys jelly, there’s a reasonable chance they’ll also purchase peanut butter and bread. But many links aren’t as obvious: perhaps people who buy apples are 12% more likely to purchase meatballs. These are the kinds of connections that association mining can uncover.

It’s easy to see how you might apply this on the business side of a grocery store.

Think about the coupons that print out at the end of a transaction. Sometimes the coupons relate to items you purchased on your shopping trip. At other times, they don’t, and you might wonder why the printer gave you 50% off frozen broccoli when you bought apples and Ziploc baggies.

Likely, the grocery store used association mining to determine that, as a person who purchased bananas, you are more likely than the average customer to buy almond milk.

Groceries

The link between these products isn’t immediately evident, but that’s why association mining is such a powerful tool. It can reveal surprising correlations that would otherwise be unavailable.

How Can You Use It?

Any organization that offers multiple products or services can use association mining.

E-commerce businesses can track grouped purchase histories in much the same way a grocery store might. They can use correlations to offer better information about what products to suggest to future customers.

Streaming media services like Netflix and Hulu use association mining to determine what recommendations to make to individual users based on their viewing histories. But any business with a website can use the technique to optimize user experience by determining what pages someone is likely to click next based on the pages they already visited.

Netflix

In healthcare, association mining can help with making diagnoses based on comorbidities. Illnesses often share common symptoms. Association mining can assign probabilities to each potential illness for each patient based on the group of symptoms they exhibit.

Applications span nearly every industry. Even if your industry isn’t listed above, your mind may be spinning with ideas on how to apply the technique at your organization.

Apply Association Mining to your Work

Association mining is possible in Python, R, or Excel with enough coding ability or knowledge of the software.

However, if you don’t want to go through all that trouble, use an intuitive, drag-and-drop data analytics platform like Rapid Insight to make association mining easy. We even have a pre-built association mining workflow available for free in the software. Just drop in your data inputs and let the software do the rest!

Whether you’re hoping to build an association mining workflow, predict customer churn, or share custom data dashboards with stakeholders at your organization, Rapid Insight makes the process easy.

On top of that, our expert analyst support team is here to guide you through your data projects every step of the way. With experience assisting hundreds of users in a wide range of industries, our analysts are on hand to help you in whatever capacity you need them.

Click the button below to schedule a demo and learn more!

REQUEST A DEMO

Stay up to date
Subscribe to our blog

Subscribe
Notify of
guest
0 Comments
Inline Feedbacks
View all comments