-
website front page
Inspiration
We are all NVCC students and given the recent string of violent crimes at Loudoun campus, we wanted a way to help NVCC students stay safe and secure during their studies at NVCC. We noticed that the crime logs for NVCC Police were public, so we already had reliable and open-source data to pull from.
What it does
SafeWalk's flagship feature is a data processing algorithm that ingests the crime log provided by NVCC Police, or any campus police department's crime log if accessible, and turns it into a Python dictionary that we can then manipulate and create visualizations for. The site then counts occurrences of crime reports per year and per hour across the campuses to provide a measure of safety. The site also implements DaVinci for a simulated chat bot that can answer questions concerning crime on NVCC campuses.
How we built it
Front end was raw HTML/CSS and JavaScript. Backend was Flask and Python.
Challenges we ran into
We attempted to implement machine learning into this project, to predict certain statistics given times of day, campus, crime classification, etc., but the hardest part of this project was actually the data processing and data science. Making sure the way we were processing our input data led to logical conclusions/results was a very large portion of this project, and had to be done before we could begin implementation.
Accomplishments that we're proud of
In the end, we were able to ingest a massive amount of semi-formatted data and then visualize it in multiple manners on a well-designed website. The data processing stage alone was a massive hurdle.
What we learned
Given a set of formatted data, the possibilities of what you can do with it are endless. There are beyond infinite ways to visualize, process, or do other things with your dataset.
What's next for SafeWalk
We would like to continue implementing our machine learning solution to expand SafeWalk into a more versatile tool for NVCC students.
Log in or sign up for Devpost to join the conversation.