Chapter 6. The user (and content) who came in from the cold

 

Open your arms and put on a big smile; it’s time to learn how to greet new customers. In this chapter

  • You’ll examine the cold-start problem that’s related to new customers.
  • You’ll learn how to segment users, so you can look at semi-personalization.
  • You’ll look at Redbubble.com as a case for cold-start problems with your newly acquired knowledge.
  • You’ll look at an implementation of a simple personalize recommender using association rules.

We’re off to a cold start in this chapter, so put on your hat and gloves and let’s get started. In the previous chapter, we talked about how to get data, and luckily, most websites have data before they start adventures into recommender systems. But even having a lot of data won’t solve the problem of how to introduce new things, be that products or users.

6.1. What’s a cold start?

Not so surprisingly, if you don’t have knowledge of your users, you can’t personalize them. And having no personalization is a huge issue because you want to make new visitors feel welcome so they’ll become loyal returning customers. Repeat customers are ideal and you’ll want to keep them happy, but there’s nothing like adding a new one to the list.

6.1.1. Cold products

6.1.2. A cold visitor

6.1.3. Gray sheep

6.1.4. Let’s look at real-life examples

6.1.5. What can you do about cold starts?

6.2. Keeping track of visitors

6.2.1. Persisting anonymous users

6.3. Addressing cold-start problems with algorithms

6.3.1. Using association rules to create recs for cold users

6.3.2. Using domain knowledge and business rules

6.3.3. Using segments

6.3.4. Using categories to get around the gray sheep problem and how to - introduce cold product

6.4. Those who doesn’t ask, won’t know

6.4.1. When the visitor is no longer new