Daily Crime Forecast
Data-Driven Crime Prevention.
Predictive Accuacy
At the time, it was twice as accurate as existing hot spot methodologies and nine times more effective than random police resource deployment.
The Opportunity
Crime patterns in urban areas can be unpredictable, making resource allocation for public safety a challenge. In 2005 Edmonton needed a data-driven tool to forecast crime trends, enabling proactive policing and improved transit safety.
The Approach
I developed the award winning Daily Crime Forecast, a predictive analytics tool that analyzed historical crime data and environmental factors to generate location-specific forecasts. These forecasts empowered officers to allocate resources strategically, enhancing safety on Edmonton’s transit system and in high-risk areas.
The Impact
The Daily Crime Forecast revolutionized how Edmonton Transit approached crime prevention, earning recognition with the 2008 NovaNAIT Challenge Award. By delivering actionable insights, the tool improved transit safety, optimized resource allocation, and demonstrated the potential of predictive analytics in public safety initiatives.
Officer Comments: It’s like fishing with a fish finder... Worked like a charm! Just turn up and wait to play.
Features
Updated Daily
Forecasts were regenerated every day to account for newest trends.
Where & When
Forecasted, down to the hour, where and when crime was likely to occur.
Fully Automated
No need for pre or post analysis, the forecasts were automatically generated from existing crime data.
Technology
- Front-end: Initially VB.NET thick client then web app: HTML, CSS, JavaScript
- Back-End: PHP, MySQL
- Analytics with Python
Disclaimer
The images presented here are for illustrative purposes only and use fictitious incidents. No real incidents or sensitive data are shown.
