Kim, Eon; (2020) Exploring the relationship between geolocated social network service text and crime. Doctoral thesis (Ph.D), UCL (University College London).

Preview

Text Thesis E.Kim.pdf Download (6MB) | Preview

Abstract

The purpose of this thesis is to explore the potential utility of Geocoded Social Network Service datasets, especially geolocated Twitter, to help understand criminal activities in urban settings. In particular, the thesis is concerned with how these datasets can advance understanding of crime events in their spatiotemporal context. Such data has potential to reveal usually unobserved underlying conditions of crime events as a complementary source of knowledge for advancing the development of crime prevention strategies. In pursuit of these goals, the thesis comprises three case studies conducted using data for New York City in the United States. It examines: 1) Mobile dynamic population patterns 2) the emotion patterns of the mobile population, and 3) the attribute patterns of the mobile population (i.e. online-traits such as type and topic of discussion). With the combined results from the three case studies, this thesis explores how adding this new type of data improves existing knowledge of crime patterns. Results demonstrated that tracking spatial-temporal fluctuations of populations, along with their emotions and concerns has potential for explaining patterns across different crime types.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONCSVXML

Archive Staff Only

View Item