Foursquare is a location based sharing application, which is one of the most popular social networks in these days. Foursquare users can share their current location, their experiences and opinions with their friends, associated with a certain venue. In this system, there is a huge amount of log data, generated by users, checking into restaurants, coffee bar, office spaces or other places. This data can be used widely for data mining tasks and analytical purposes.
In my thesis, I describe the published, and the potential researches related to this field, then I introduce the freely available Foursquare data sets, and the data collector module, implemented by myself.
One of the mentioned tasks is defining different behavior patterns of Foursquare users, building data mining models on these patterns, and predicting check-in activities in the future, using machine learning methods. My task was to create a system with this functionality based on the accessible but incomplete data sets.
Keywords: Foursquare, LBSN, data mining, machine learning, clustering, classification