Tools for Videogame Discovery Built Using Latent Semantic Analysis

TitleTools for Videogame Discovery Built Using Latent Semantic Analysis
Publication TypeJournal Article
Year of Publication2015
AuthorsRyan, JOwen, Kaltman, E, Mateas, M, Wardrip-Fruin, N
JournalProceedings of the 10th International Conference on the Foundations of Digital Games
Date PublishedJune
KeywordsEIS, computational linguistics, gamenet, gamesage, latent semantic analysis, lsa, machine learning, natural language processing, o game discovery, tools, visualization
We introduce two tools for videogame discovery built using latent semantic analysis, a foundational technique in statistical natural language processing. The first, GameNet, is an explorable network in which 11,829 games are linked according to how related they are. The second tool, GameSage, takes a user's free-text description of a game (or an idea for a game) and integrates with GameNet to present an interactive listing of the existing games that are most related to the game she described. Both tools are hosted online as web apps.