As part of Steam Labs experiments, Valve recently rolled out three projects that aim to improve visibility of games on Steam. GabeN and Co now shed more light on Interactive Recommender, which should make searching for games much easier.
"One of Steam's strengths is its massive catalog of great games from developers large and small, spanning almost every genre. With so much stuff to choose from, we've heard from users that you'd like better tools to help you find games you'll enjoy", Valve said.
Steam's Interactive Recommender relies on machine learning to sort the games and the algorithm gives personalised recommendations to players based on their playing patterns.
Valve added real-time controls to fine-tune Interactive Recommender's results, which should be convenient for times when you're looking for something completely different from what you normally play.
"Underlying this new recommender is a neural-network model that is trained to recommend games based on a user's playtime history, along with other salient data. We train the model based on data from many millions of Steam users and many billions of play sessions, giving us robust results that capture the nuances of different play patterns and covers our catalog", they said.
Steam users can restrict the results by release date and even popularity, or "mainstream-ness" as Valve called it, which should be awesome news for indie developers, who have found themselves on the wrong end of Steam's visibility-inducing algorithms too many times lately.
Interactive Recommender's results will be completely personalised and based on your playing habits rather than external info, with release dates being the only exception. Neither reviews nor tags alone can affect the results, which technically renders review-bombing irrelevant.
Steam users will still be able to narrow the final results by tags, but they're not part of Valve's deep-learning model.
Given Interactive Recommender's personalised nature, the feature doesn't require any work on the behalf of developers. They will, however, be able to check how many page visits they owe to Steam's new feature via Traffic Breakdowns.
You can find Valve's blog post on Steam's Interactive Recommender here.