Social media data for content creation in location-based games

Abstract

Content in location-based games (LBGs) fundamentally depends on the location, and its global coverage is challenging. Typical content creation approaches involve crowdsourcing, but recent research has also considered Web crawling and OpenStreetMap data as potential resources. A real-world location is a fundamental element in LBGs content. Media items such as images, short video clips, names, geocaches, and riddles can also be attached to the locations for games supporting sightseeing tours and treasure-hunts. In this paper, we evaluate the usability of social media data for content creation purpose in LBGs. We focus on three content items: location, image, and name. For this purpose, we retrieve geotagged images from services such as Flickr, Foursquare, Yelp, and Google Places from 2019 to 2022. Our experiments with six regions show that the Flickr images are representative of place with higher probability (32%) and location accuracy (96%), but relevant name extraction relied on the external method. Foursquare and Yelp images always have the correct location (100%), and the name is usually relevant (100%), but the image is rarely representative (22%, 4%). Google Places, despite providing large volumes of data and a variety of establishments, only 5% of representative images were found; however, location accuracy (75%) and relevant names (100%) are reasonable measures.