Abstract:
The popularity and diffusion of Social media have been growing constantly in the last years, making the automatic understanding the giant amount of data produced fundamental in discovering recurrent patterns and other important information. While a huge body of work can be found in the literature on the topic of extracting "mood" information about a topic from textual information, very little work has been done on the problem of automatically analyzing the visual content of images in social media.
In this thesis the images retrieved from social media are used to analyze how Venice is represented in touristic photographs in different times of the year and in its different areas (sestieri). To this end, using techniques mutuated from the object classification literature, we built a classifier able to distinguish new photos' category, and analyzed the variation of class distribution in space and time, thus providing an quantitative characterization of the visual narrative of Venice in social media.
