There are a lot of games on Steam. What the heck do you want to play? Valve is hoping to solve that particular problem with machine learning as the new interactive recommender is launched on Steam, a machine learning system that records your preferences, compares them to those of other players with similar histories, and recommends titles you may not have heard of based on all of those settings. So you can find titles you may never have even been aware of in the first place.
It’s important to note that rather than basing its evaluations on reviews or the like, the metric is entirely based on playtime. Essentially, the system is looking at your playtime, comparing it to others, and looking for games that others have played but you have not and recommending it thusly. While that doesn’t mean it’ll hit it out of the park, it’s worth playing around with it just the same to see if it might uncover some unexpected gems.