WYLDFIRE

Working
Model
Homepage

Inspiration

During the few years, terrible wildfires have spread all over California, Australia, and all over the world. The images and videos are heartbreaking. When the summer season starts, these events tend to appear more often, and the final results are catastrophic. This is why I decided to use AI to solve this real world problem and be able to help those people who are affected by these fires In this article, I will guide you to a step-by-step tutorial about predicting the spread of wildfires, using AI.

What it does

Given that a wildfire is raging nearby, this model tries to predict whether there will be a wildfire in the user's city in the upcoming days and whether or not he should evacuate

How I built it

First, the wildfires dataset was taken from Kaggle. Then the model was built by passing the data through a ridge classifier. Finally the ml app was hosted on Streamlit and then integrated with HTML/CSS

Challenges I ran into

Finding the most accurate ML model for prediction and then deploying ml app on streamlit

Accomplishments that we're proud of

The ML model predicts whether or not the user is in danger of wildfire with 75% accuracy and gives him the expected duration of the wildfire reaching his city. This seems to be a promising path in the wildfire prediction challenge the world faces