Mining geo-referenced community-contributed multimedia data

Abstract

© 2012 Springer-Verlag London. All rights are reserved. Besides connecting users and allowing interactions between them, social networks are becoming an increasingly popular medium for sharing multimedia content, such as images and videos. Due to technological advances it has become extremely simple to create and share such content in (near) real-time, and even associate it with a location where it was made (i.e. geo-reference it). All of this has caused tremendous amounts of geo-referenced multimedia content to become publicly available, which made it suitable for analysis by employing different visualization and data-mining techniques. This chapter presents some of the techniques and methods for mining geo-referenced multimedia content in order to discover patterns and trends in it, which can lead to better understanding of the phenomena driving the data generation in the first place.