Abstract

How is it possible to use social networks to gain better understandings about the emotional life of cities? Is it possible to use data generated by city dwellers on social networks to understand whether certain spaces and places in the city are the object of systematic emotional expressions, in certain times of the day, week, month or year, and across different contexts? City dwellers have radically transformed the ways in which they relate, work, entertain themselves, consume, produce and express themselves and their emotions. This is due to multiple reasons, across the transformation of the dimensions of time, of the new forms of labor and of production, of the new structures and organizational models in society, and with the emergence of novel modalities and tools for communication, interaction, information, knowledge. In our research we have gathered large amounts of content generated on social networks by city dwellers of different kinds (citizens, tourists, from different communities and speaking one of the observed 29 languages) in the cities of Rome, Milan, Turin, Berlin, Sao Paulo, Montreal, Toronto, New York, New Haven, Hong Kong, Cairo and Istanbul. We analyzed this data through Natural Language Processing to highlight, wherever possible, the emotions expressed in each public message, and its relation to a geographical location within the city. Results indicate how, in certain cases, it is possible to describe what we have called "Emotional Landmarks": dynamic landmark locations which temporarily or recurrently host systematic emotional expressions, in large quantities and in recurrent, sometimes predictable ways.