Researchers at the University of California, Santa Cruz Center for Games and Playable Media have created a playable visualization of 16,000 video games grouped according to shared commonalities such as gameplay, cultural reactions, and game designers.
The three-dimensional (3D) GameSpace visualization resembles a constellation of stars, and players can navigate through clusters of related titles.
For each game, users read its Wikipedia article or watch a YouTube video of gameplay. Game data is pulled from all of the articles about specific games found on Wikipedia. Researchers then use latent semantic analysis, a natural language-processing technique, to analyze the game descriptions and compute similarity scores. The scores are used to determine the position of each game within the 3D galaxy.
GameSpace will continue to grow as each night the system will search Wikipedia and add any new game articles to the visualization.
From UCSC NewsCenter
View Full Article
Abstracts Copyright © 2017 Information Inc., Bethesda, Maryland, USA
No entries found