Urban communities can benefit from behavior regulation of their members in the interest of collective values. The absence of such control is related to the concept of social disorganization and is hypothesized to be associated with crime and anti-social behavior in neighborhoods. Social disorganization is, however, hard to quantify due to the lack of data and the inherent complexity that emerges from social interactions. Notably, geolocated social media provides a real-time assessment of places via the examination of the digital footprints left by users. In this paper, we introduce a measure for social disorganization by analyzing geotagged posts on Twitter. We propose to characterize the social disorganization of a place by evaluating the entropy of individuals’ opinions about certain subjects. As a case study, we used tweets related to football in the UK, given its ubiquity in that country, which makes its supporters as proxies for the social characteristics of those places. We found that our proposed measure can reasonably explain the variation of the occurrence of crime across regions in UK and that our measure better explains the variation of crime among places with higher social disorganization.
AI4BigData at the 30th International FLAIRS Conference, 2017, Marco Island, Florida, USA.
[Download preprint here]
[Link to publication]