Predicting venues in location based social network

Abstract

The circulation of the social networks and the evolution of the mobile phone devices has led to a big usage of location based social network applications such as Foursquare, Twitter, Swarm and Zomato on mobile phone devices, which signifies a huge dataset containing a blend of information about users behaviors, social society network of each users and also information about each of venues. All this information is available in mobile location recommendation systems. These datasets are much more different from those which are used in online recommender systems; besides, they have more information and details about the users and the venues allowing to have more clear results with much more higher accuracy of the analyzing in the results. In this paper we examine the user’s behaviors and the popularity of the venue through a dataset with large check-ins from a location based social services, i.e. Foursquare, by using large scale dataset containing both user check-in and location information. Our analysis exposes across 3 different cities. The analysis of this dataset reveals a different mobility habits, preferring places and also location patterns in the user personality. The information about the user’s behaviors and each of the location popularity can be used to know the recommendation systems and to predict the next move of the users, depending on the categories form which the users attend to visit, according to the history of each user’s check-ins.