Empirical Observation of User Activities: Check-ins, Venue Photos and Tips in Foursquare

Abstract

Location-based social networking platform (e.g., Foursquare), as a popular scenario of participatory sensing system that collects heterogeneous information (such as tips and photos) of venues from users, has attracted much attention recently. In this paper, we study the distribution of these information and their relationship, based on a large dataset crawled from Foursquare, which consists of 2,728,411 photos, 1,212,136 tips and 148,924,749 check-ins of 190,649 venues, contributed by 508,467 users. We analyze the distribution of user-generated check-ins, venue photos and venue tips, and show interesting category patterns and correlation among these information. In addition, we make the following observations: i) Venue photos in Foursquare are able to significantly make venues more social and popular. ii) Users share venue photos highly related to food category. iii) Category dynamics of venue photo sharing have similar patterns as that of venue tips and user check-ins at the venues. iv) Users tend to share photos rather than tips. We distribute our data and source codes under the request of research purposes (email: yi.yu.yy@gmail.com).