Forest Fire Prediction

Every year, devastating forest fires ravage vast areas of our planet, leaving behind a trail of destruction and environmental damage. The urgency to predict and prevent these catastrophic events has never been more critical. In the pursuit of a proactive approach, we turn to the power of machine learning regression to predict the probability of forest fire occurrences. By harnessing the potential of predictive modeling, we aim to mitigate the impact of these natural disasters and safeguard our precious ecosystems.

Objective of this project:

Primary Objective:

Forest fire prediction aims at predicting the probability of the occurance of forest fire based on 3 input parameters. These parameters are Oxygen content in the air, Humidity in the atmosphere and the temperature.

Secondary Objective:

• Use of Logistic Regression Machine Learning model
• Use of Pickel library
• Use of Flask module for frontend development
• Developing a user friendly Frontend UI

Libraries used:

• Pandas: Used to read, clean and manipulate data from csv file.
• NumPy: Used to create easy to use and flexible array objects.
• Flask: Use the flask module to link the machine learning model at the backend to the frontend.
• Pickel: To serialize and deserialize the data to be sent between the frontend and the backend.
• Sklearn: Used for the logistic regression model.

To install the required libraries run the following code in the terminal of your project environment:

pip install pandas numpy flask pickel sklearn