All the UTPD incident emails that came too late. As students tired of hearing about criminal incidents near campus way after they happened, we wanted to do something to change that.

What it does

Allows witnesses to post crimes they've witnessed, along with a video and description. The video and description are then analyzed with machine learning to verify the accuracy of the claim and to provide more insights to law enforcement. We also allow students to view a map -interface that shows the locations of current/previous crimes as well as estimated future hotspots to avoid. Using this data we can generate PowerBI visuals.

How I built it

After thoroughly analyzing publicly available APD crime data, we build SaferAcres. SaferAcres is a full-stack web application built with front and back-end functionalities. We used Azure's Machine Learning studio to forecast future criminal activity, stored data in Azure's SQL databases, analyzed crowdsourced data with Azure's cognitive functions, and build a robust backend with Python's Flask library.

Challenges I ran

We struggled with the design and creation of well-structured databases with Azure. There was also the challenge of building an entire full stack application in 24 hours.

Accomplishments that I'm proud of

We were able to use Azure's Machine Learning studio to forecast future crime locations. Additionally, users are able to interact with historical crime data, live crowdsourced data, and criminal activity predictions simultaneously.

What's next for Safer Acres

Integrating with student services like Sure Ride and UTPD

Share this project: