Machine Learning-Powered Steam Recommendations Now Live
Steam's latest feature will recommend you games not only based on what you've played, but on what other people will similar taste have played.
Steam has added a new tool today for recommending games to users. Steam Interactive Recommender uses machine learning to predict which games you'll enjoy based on what other people with similar taste have played.
The Interactive Recommender started as a tool within Steam Labs back in July 2019, but has now become a proper feature of the store.
The announcement digs into how all of this works. You can interact with the recommender by specifying that you want popular or niche games, or whether you want older or newer titles, and it'll give you a list of titles based on other Steam users' playtime histories.
Tags and reviews are not taken into account, and it's based purely on what people are actually playing. "The basic idea is that if there are other players with similar play habits to you, who also play a game that you haven't tried yet, then that game is likely to be one you'll enjoy too," the announcement reads.
Judging on my own taste, it seems very accurate--it recommended me several games that I have played and enjoyed on consoles.
The Interactive Recommender isn't replacing any of Steam's other recommendation methods, but will exist alongside them.
If you're looking for more recommendations for new things to play, check out our list of games to play while you're stuck inside.
The products discussed here were independently chosen by our editors. GameSpot may get a share of the revenue if you buy anything featured on our site.
Got a news tip or want to contact us directly? Email firstname.lastname@example.org
Join the conversation