James Ryan, Eric Kaltman, Andrew Max Fisher, Taylor Owen-Milner, Michael Mateas, and Noah Wardrip-Fruin
07/18/2017 09:55 AM
Computational Media
We present GameSpace, a visualization of the videogame medium as a 3D explorable space. GameSpace takes the form of a galaxy of 15,000 stars, where each star represents an existing game and stars are positioned such that related games are nearer to one another. Users may fly around the space, using conventional game controls, to explore the medium and serendipitously discover new games. This application is made possible by a combination of techniques from natural language processing and machine learning, namely latent semantic analysis and multidimensional scaling, and builds upon our earlier work in producing tools for videogame discovery. Beyond demonstrating an increase in user engagement relative to our primary existing tool, in this paper we outline our design goals, describe our development process, and discuss the project along various dimensions of intrigue. GameSpace is available online at http://gamespace.io.
