Skip to content
Article

Conquer the world with Azure Machine Learning

What a great idea! You’ve decided to convert your brick and mortar store into an online supermarket. You’ll create intelligent business processes that will make shopping easy. They will allow you to run the web shop efficiently! You’ll make Amazon and Ali Baba jealous. Let's investigate how using Azure Machine Learning can help you.

Creatures of habit

Most people regularly buy the same items. For example, imagine a customer who regularly buys some cheese. It would be nice if the system would advise him to put it in his shopping cart. But he probably won't need cheese during each visit to the shop, only when he has run out. So the system should be smart enough to figure out if he needs it. If you repeat this process for all of his regular products, you can produce an automatically generated grocery list. Which leads to one-click shopping! Now how's that for a time saver?

Out of cheese?

Azure Machine Learning (ML) provides a way to answer this question. First you need to identify the type of problem you need to solve. Your problem has two possible outcomes: either your customer will buy cheese, or he won’t. Because the possible outcomes are limited to just two categories here, this problem describes what is called a Binary Classification. Fortunately Azure ML provides some great out-of-the-box classification algorithms to solve these kinds of problems. You use a binary classification algorithm to make predictions based on past data. For instance, the fact that a certain customer buys cheese every week will stand out as a pattern. Azure ML can detect this pattern and predict whether the customer will buy some cheese when he’s in your web shop.

Explore more articles