Public Perceptions on Organised Crime, Mafia, and Terrorism: A Big Data Analysis based on Twitter and Google Trends

Public perceptions enable crime and motivate government policy on law and order; however, there has been limited empirical research on serious crime perceptions in social media. Recently, open source data-and 'big data '-have enabled researchers from different fields to develop cost-effective methods for opinion mining and sentiment analysis. Against this backdrop, the aim of this paper is to apply state-of-the-art tools and techniques for assembly and analysis of open source data. We set out to explore how non-discursive behavioural data can be used as a proxy for studying public perceptions of serious crime. The data collection focused on the following three conversational topics: organised crime, the mafia, and terrorism. Specifically, time series data of users' online search habits (over a ten-year period) were gathered from Google Trends, and cross-sectional network data (N=178,513) were collected from Twitter. The collected data contained a significant amount of structure. Marked similarities and differences in people's habits and perceptions were observable, and these were recorded. The results indicated that 'big data' is a cost-effective method for exploring theoretical and empirical issues vis-a-vis public perceptions of serious crime.