Mining Social Images to Analyze Routing Preferences in Tourist Areas

Abstract

Social media platforms provide a useful source of data for environmental planning. In the last years these data have been exploited to perform social behaviour analysis. This work uses the huge amount of georeferenced images publicly available on social media as a source of information to infer the behaviour of tourists. Visual analytic mapping tools combined with the Parzen-Rosenblatt non-parametric kernel density estimation give us visual clues to assess the attractiveness of tourist geographical areas. To investigate the preferred combinations of locations visited by the tourists within a time window of few days we propose to mine association rules using the Apriori algorithm. A prototype of an integrated system to visually perform the suggested analysis has been realized and the paper reports about some of case studies performed with it.