Award Information
Description of original award (Fiscal Year 2023, $5,000)
This entry proposes to use data from Airbnb property listings and customer reviews, particularly those classified under the "location" subcategory, to assess and quantify levels of fear of crime within communities. Entrants propose the application of computational methods, with a primary reliance on Large Language Models (LLMs), to extract neighborhood-specific information from the Airbnb dataset. The proposed text analysis process includes text classification, sentiment analysis, and topic modeling. Both supervised and unsupervised LLM-based text mining techniques would be used. This proposal illustrates how an existing data set can be used to measure community fear of crime.
The lone team member for this submission was Ruilin Chen.
This entry was one of two first prize winners in the category 2, fear of crime, construct competition.
Similar Awards
- Spatial analysis of social vulnerability and crime disparities through interpretable machine learning
- NIJ Innovations in Measuring Community Perceptions Challenge Winner: Duddon Research
- NIJ Innovations in Measuring Community Perceptions Challenge Winner: Policing Accountability and Policy Evaluation Research Lab