News Releases

September 6th, 2017

YOKOZUNA data wins the two tracks of the IEEE Game Data Mining competition organized by NCSOFT

Tokyo, Japan, (September 6, 2017) – Middleware and game development company Silicon Studio Corporation is proud to announce that their new data science platform, YOKOZUNA data, has won the Game Data Mining competition at the IEEE 2017 Conference on Computational Intelligence in Games (CIG), held in New York. The contest was based on usage of player datasets from NCSOFT’s hit game Blade & Soul.

Yokozuna Data

Competing with both academics and companies from all around the world, YOKOZUNA data emerged as the best predictor, being able to accurately predict not only which players were going to leave the game soon, but also the exact moment when they were going to do it; placing YOKOZUNA data, Silicon Studio and Japan at the forefront of Data Science.

Annually CIG conference hosts a variety of competitions. This year the Game Data Mining contest, was based on datasets utilizing actual player data from the highly successful game Blade & Soul, developed by NCSOFT – one of the world’s largest developers of massive multiplayer online role-playing games (MMORPGs). The contest judged participants on the following two tracks:

  • Churn prediction track: A prediction challenge to indicate which players were going to quit the game.
  • Survival analysis track: The goal being to predict the lifetime of each player or how long each player would play the game.

YOKOZUNA Data, a market-leading machine-learning engine designed to predict the behavior of individual players, won both tracks. The contest served to benchmark Silicon Studio’s solution against the brightest minds in gaming research, demonstrating the advanced capabilities of YOKOZUNA Data. This win validates the advanced abilities and widespread range of YOKOZUNA Data as being able to produce results of the highest-quality not only for mobile games, but also for MMO games.

IEEE 2017 Conference on Computational Intelligence in Games

Official competition website


Blade and Soul

About YOKOZUNA data

YOKOZUNA data is a state-of-the-art machine-learning engine to predict individual player behavior. It consists of a recommendation system and a player prediction platform that utilize next generation artificial intelligence algorithms to move game development into the future.

About Silicon Studio

Silicon Studio is a Japan-based game engine and middleware company providing quality rendering, optical effects, and post-effects solutions for game development. Silicon Studio also develops and publishes games across mobile, PC and consoles, and develops games for third parties.

All other names and trademarks mentioned are the registered trademarks and property of the respective companies.

Have questions?

Find answers to frequently asked questions, request evaluation versions, or ask us any other questions from here.