AntiDote
Description:
We predict the role of fentanyl in opioid deaths. We are interested in how the opioid crisis is currently affecting America.
Inspiration:
The opioid crisis has swept America and we were interested in the data behind the madness.
What it does:
We apply machine learning algorithms in order to predict if fentanyl will result in an overdose.
How we built it:
All of out back-end code is handled in Python. The front end of the application was created using HTML, CSS, Javascript and Jinja. The data from the back-end is sent to the front end via render calls in flask. The variable data is then output to HTML for the user to see.
Challenges we ran into:
We had difficulty choosing the right algorithm to make predictions. At first, we attempted an implementation of KNN, but the variance in the data, as well as a failed attempt at normalization made it quite difficult.
Accomplishments that we're proud of:
We are proud of bring able to implement a working machine learning algorithm. We are also happy to to create a webapp in python, which we have never done.
What we learned:
We learned how to use build web apps in python using flask, apply machine learning algorithms to data sets, and how to integrate Javascript into a python flask project.
What's next:
We need to get a hold of real-time data in order to predict upcoming overdoses.
Built with:
Python, Flask, Javascript, HTML, CSS, Pandas, Numpy, SciPy, Jinja
Prizes we're going for:
Call of Duty: Black OPS 4 (XBOX ONE)
$100 Amazon Gift Cards
Social Entrepreneurship Award
Hustle Award
Grand Prize
Jetbrains Pro Software
Misfit Shine 2
Fujifilm Instax Mini 26