Jiahui Wu’s Dissertation Defense

Event Start Date: Friday, June 25, 2021 - 11:00 am

Event End Date: Friday, June 25, 2021 - 1:00 pm


Crime imposes significant costs to society at individual, community, and national levels. Reported crime statistics are important in quantifying the severity of crimes, based on which decision-makers would allocate resources for crime interventions, such as predictive policing. With the advent of information and communication technology, the big data of human mobility has triggered the interest from researchers in various fields to study the relationship between urban crimes and mobility at a large scale, especially the predictive power of urban mobility for urban crimes. This research direction can enrich our understanding of crimes and better informed crime-related decision-making. One big concern about measuring crimes with reported crime statistics is its bias issue. The bias could be produced by different levels of residents’ willingness to report potential crime incidents, the various degrees of police activity in neighborhoods, and even the bias of the justice system to decide whether an incident should be recorded as a crime incident. While lots of studies about crime prediction are aware of biases in reported crime data, few of them propose solutions to address or mitigate this issue or to evaluate how this issue would affect their prediction models in terms of accuracy or fairness.

My dissertation research aims to explore the potential of human mobility big data for crime prediction. Specifically, my dissertation will advance the state of the art by conducting three strands of empirical studies that address three challenges in mobility-based crime prediction:

  1. The construction of mobility-based features might be sensitive to different methodological choices. One critical area of mobility behavior analysis to predict crime is the identification of urban hotspots. There are various methodological decisions to identify and quantify hotspots in cities. There might be conflicting findings without careful examination of these choices. Therefore, my work performs a systematic spatial sensitivity analysis on the impact of these methodological choices, provides guidelines to identify the most stable ones, and explores the impact of methodological sensitivity on crime prediction tasks.
  2. Under-reporting generates data bias in reported crime data. To address such bias, my work develops a Bayesian statistical model for long-term crime prediction that infers the unobserved true number of crime incidents. Comprehensive experiments show how the accuracy and fairness of long-term crime prediction would be affected by modeling the under-reporting of crimes.
  3. Although empirical studies show promising results about the relationship between human mobility and long-term crime prediction, the effects of mobility features on short-term crime prediction have yet to be explored. Therefore, my work conducts a series of experiments to explore how incorporating mobility features into short-term crime prediction models affects their performance in terms of accuracy and fairness.


  • Associate Professor Vanessa Frias-Martinez, Chair
  • Professor Kathleen Stewart, Dean’s representative
  • Professor Richard Marciano
  • Assistant Professor Wei Ai
  • Assistant Professor Taylor M. Oshan