Inspiration
We were thinking of a way to alert people in Australia about the bush and forest fires. We wanted to develop an alert system where they could see how much danger they are in, with respect to the location of fires near them.
What it does
The app uses your location and temperature to determine if your location could be an early onset location for fires and sees how dangerous the neighboring fires are from you.
How I built it
We built the backend using a K-Nearest Neighbors algorithm in python. Then the front end was built with HTML/CSS and Javascript utilizing weather api's and google maps API.
Challenges I ran into
Finding good data was hard because the incident was so recent. It took some time to find the data set I used. The other difficulty was getting the whole front end to cooperate with the backend.
Accomplishments that I'm proud of
Building our own machine learning algorithm without using an API. Able to make the front end and backend communicate seamlessly.
What I learned
We learned how to properly integrate various different components of an application into one cohesive interface.
What's next for FireWatch
We hope to improve more on the UI and hopefully add more features to the machine learning model.
Log in or sign up for Devpost to join the conversation.