This is going to draw on a few insights. For one thing, we talked about people being fickle-their tastes can change over time, so comparing one person to another person based on their past behavior becomes pretty complicated. People have different phases where they have different interests, and you might not be comparing the people that are in the same phase to each other. But, an item will always be whatever it is. A movie will always be a movie, it's never going to change. Star Wars will always be Star Wars, well until George Lucas tinkers with it a little bit, but for the most part, items do not change as much as people do. So, we know that these relationships are more permanent, and there's more of a direct comparison you can make when computing similarity between items, because they do not change over time.

The other advantage is that there are generally fewer things that you're trying to recommend than there are people you're recommending to. So again, 7 billion people in the world, you're probably not offering 7 billion things on your website to recommend to them, so you can save a lot of computational resources by evaluating relationships between items instead of users, because you will probably have fewer items than you have users in your system. That means you can run your recommendations more frequently, make them more current, more up-to-date, and better! You can use more complicated algorithms because you have less relationships to compute, and that's a good thing!

It's also harder to game the system. So, we talked about how easy it is to game a user-based collaborative filtering approach by just creating some fake users that like a bunch of popular stuff and then the thing you're trying to promote. With item-based collaborative filtering that becomes much more difficult. You have to game the system into thinking there are relationships between items, and since you probably don't have the capability to create fake items with fake ties to other items based on many, many other users, it's a lot harder to game an item-based collaborative filtering system, which is a good thing.

While I'm on the topic of gaming the system, another important thing is to make sure that people are voting with their money. A general technique for avoiding shilling attacks or people trying to game your recommender system, is to make sure that the signal behavior is based on people actually spending money. So, you're always going to get better and more reliable results when you base recommendations on what people actually bought, as opposed to what they viewed or what they clicked on, okay?