An emerging perspective in criminology suggests that we can better understand where crime occurs by considering the characteristics of building owners and place managers. To get a better understanding of how commercial businesses are managed in relation to crime, I have collected Yelp reviews from 30 American cities. Using online reviewers as 'eyes on the 'street', I am analyzing these reviews using natural language processing and machine learning techniques to investigate how management characteristics relate to crime and to explore how the relationship between managers and crime changes across urban environments.

While online-based research often analyzes content to understand the people behind the keyboards, there are also increasing opportunities to learn from the crowd wisdom of online "experts". To this end, I am leading a team that is conducting research on gangs through an online messageboard for fans of Chicago rap and gang culture. Through natural language processing strategies, we have trained AI to read the message board and have generated a social network of gang-affiliated people associated with Southside gangs. By comparing our data to official records and by replicating analyses conducted with other sources of data on gangs, we can show that our crowd-sourced dataset is a valid source for scholarly research. Because research on gangs is often limited to scholars who collaborate with public officials, generating crowd-sourced data from publicly available content can allow new voices to join the conversations around gang violence and criminal justice policy.