Mining Frequent Spatio-Temporal Patterns from Location Based Social Networks

Location Based Social Networks (LBSN) like Twitter or Instagram are a good source for user spatio-temporal behavior. These social network provide a low rate sampling of user’s location information during large intervals of time that can be used to discover complex behaviors, including frequent routes, points of interest or unusual events. This information is important for different domains like route planning, touristic recommendation systems or city planning. Other approaches have used the data from LSBN to categorize areas of a city depending on the categories of the places that people visit or to discover user behavioral patterns from their visits. The aim of this paper is to analyze the frequent spatio-temporal patterns that users share when visiting a city. This behavior is studied in a well limited geographical area by means of frequent itemsets algorithms in order to establish some causal dependence between visits that can be interpreted as interesting routes or spatio-temporal connections. The data analyzed was obtained from the public data feeds of Twitter and Instagram inside the area of the cities of Barcelona and Milan for a period of several months. The analysis of these data shows that these kind of algorithms can be successfully applied to data from any city (or general area) to discover useful patterns that can be interpreted on terms of the city singular places and areas and that these patters can be used as a the elements of a knowledge base for different applications.