In 2016,Niantic Labs released Pokémon Go,an augmented reality smartphone game that attracted millions of users worldwide.This game allows users to“catch”Pokémons through their mobile cameras in different g...In 2016,Niantic Labs released Pokémon Go,an augmented reality smartphone game that attracted millions of users worldwide.This game allows users to“catch”Pokémons through their mobile cameras in different geographic locations that often correspond to prominent places.This paper analyzes the distribution of PokéStops,Pokémon gyms,and spawnpoints in selected urban areas of South Florida and Boston.It identifies which socioeconomic variables and landuse categories affect the density of PokéStops,and how PokéStops and gyms cluster relative to each other.Using nearest neighbor analysis,this paper assesses also how actual PokéStop locations are reflected in Yelp’s“PokéStop nearby”attribute.Results show that black and Hispanic neighborhoods are disadvantaged when it comes to crowd-sourced data coverage,that PokéStops occur more frequently in commercial,recreational and touristic sites and around universities,and that PokéStops tend to cluster around gyms.The latter suggests that these point sets were generated by a similar location selection process.To mitigate geographically linked biases,future versions of augmented reality and geo-games should aim to make them equally accessible in all areas,for example by placing extra resources,such as points of interest,in neighborhoods that are currently underrepresented in data coverage.展开更多
文摘In 2016,Niantic Labs released Pokémon Go,an augmented reality smartphone game that attracted millions of users worldwide.This game allows users to“catch”Pokémons through their mobile cameras in different geographic locations that often correspond to prominent places.This paper analyzes the distribution of PokéStops,Pokémon gyms,and spawnpoints in selected urban areas of South Florida and Boston.It identifies which socioeconomic variables and landuse categories affect the density of PokéStops,and how PokéStops and gyms cluster relative to each other.Using nearest neighbor analysis,this paper assesses also how actual PokéStop locations are reflected in Yelp’s“PokéStop nearby”attribute.Results show that black and Hispanic neighborhoods are disadvantaged when it comes to crowd-sourced data coverage,that PokéStops occur more frequently in commercial,recreational and touristic sites and around universities,and that PokéStops tend to cluster around gyms.The latter suggests that these point sets were generated by a similar location selection process.To mitigate geographically linked biases,future versions of augmented reality and geo-games should aim to make them equally accessible in all areas,for example by placing extra resources,such as points of interest,in neighborhoods that are currently underrepresented in data coverage.
