Detecting attractive locations and tourists' dynamics using geo-referenced images

This paper presents an approach to detecting locations attractive to tourists and discovering tourists' dynamics based on data obtained from publicly available geo-referenced images. Results of an analysis based on geo-visualization methods and data obtained from Flickr are presented in this paper. We focus on data tagged as recorded in Berlin, Germany. The goal is to identify attractive locations in the city, assuming that an attractive place is characterized by large amounts of photos taken by many people. In addition, we attempt to identify standard routes people take when making a tour. To achieve this, we propose an approach based on density-based clustering and route similarity and dynamics clustering. The results presented in the paper indicate that information obtained by analyzing Flickr photos can be used to reliably detect locations attractive to tourists and suggest best tour routes.