Gamasutra is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Gamasutra: The Art & Business of Making Gamesspacer
View All     RSS
September 18, 2019
arrowPress Releases

If you enjoy reading this site, you might also want to check out these UBM Tech sites:

Press Releases
  Machine learning now understands how the game looks and plays
  Share on Twitter RSS

[This unedited press release is made available courtesy of Gamasutra and its partnership with notable game PR-related resource GamesPress.]

Vilnius, May 23, 2019 - Games media service RAWG launches a neural network to help you find a new favorite game in a few seconds. RAWG brings a discovery-rich list similar to popular content discovery platforms, that is powered by the world's largest games database, and lets you browse an endless flow of games. If you like what you see, our neural network will find more games you will enjoy in a single click.

“Analyzing people’s search scenarios, we noticed that the most frequent request is to find ‘Games like X’. Our neural network does just that. It analyzes in-game content along with a pinch of metadata and heuristics to understand how the game looks and plays, and matches it with other games from any platform you game on.” said Gadji Makhtiev, the founder of RAWG.

Metadata-based recommendations rely heavily on tags and categories, which makes them less accurate for multi-platform gamers. Machine learning solves that, as the neural network can analyze all games as soon as they are in the database and use something better than the metadata—the actual in-game content.

“Machine learning recommendations work differently than those based on metadata or collaborative filtering. If you liked a game like Super Mario Galaxy, you'd probably be looking for other cheerful colorful games and not just 3D platformers. If you enjoyed a sports game like EA Sports NHL, chances are you are going to like more sports games or management simulators. If you know what you are looking for, RAWG will help you find your next favorite game in less than 20 seconds.” said Gadji.

How games discovery works: You can follow genres, platforms, creators, and your friends, and monitor game recommendations that are relevant to them in one place. If you like what you see, you can click on the game card where you will instantly find stores to buy it. The “Show more like this” button will take you to a page of similar games. The results you see there are provided via machine learning. The neural network analyzes in-game content of all the games in the database and finds the most similar ones.


About: RAWG consists of the biggest games database in the world (over 316,000 titles) and a service to keep track of your library across all devices. It strives to create a central place for gamers on all platforms to discover something new and share their gaming experiences. RAWG is a startup based in Lithuania, and it is already out of beta on