Forest Fire Prediction

Inspiration

1 billion dollars damage, 7000 buildings were destroyed, 42 people died from the wildfire in northern California in late 2017. Forest fire is no longer some irrelevant natural disaster.

What it does

This is a prototype for early detection of forest fire based 8 environmental factors: FFMC, DMC, DC, ISI, wind, rain, temperature, Relative Humidty.

How I built it

As a team, we built it with matlab's machine learning to classify the data and designed a user interface using app designer.

Challenges I ran into

Well, we learnt matlab from scratch and learnt new syntax, function, formula.... The dataset was difficult to collect too. Furthermore, when we were writing, we could not get the app designer on matlab to speak to the workspace.

Accomplishments that I'm proud of

We are so proud that we started from scratch and made a prototype of our original idea.

What I learned

matlab. Believe in yourself. Work as a team.

What's next for Forest Fire Prediction

This is the initial stage of Forest Fire Prediction. We believe that with more datasets, the accuracy of the prediction will improve. Forest Fire Prediction 1.0 displays a binary answer, in the future we look to predict probability of forest fire occurrence. After numerous updates, we hope that the app can alert the government at an early stage of forest fires. In this way, the government can help reduce loss caused by the forest fire.

Built With

• matlab