In this release, we have finalized all features and conducted extensive testing to ensure the reliability and accuracy of our predictions. Here are the key components of this release:
Machine Learning Approach: We have implemented a Random Forest machine learning algorithm to predict deforestation patterns in the Paraguayan Chaco over the next decade. This algorithm is capable of managing large datasets with a multitude of variables, making it ideal for our complex task.
Datasets: We have utilized several datasets integral to the analysis and subsequent predictions of deforestation patterns. These datasets provide a historical context for land use and its potential impact on deforestation.
Training and Validation: We have employed the Balanced Random Forest Classifier from the imblearn library for our model. The model has been trained and validated using a comprehensive set of features, enhancing its predictive accuracy and reliability.
Prediction: The predictive model, trained and validated on data from the developed region, is now applied to the undeveloped region of the Paraguayan Chaco. The model provides a nuanced understanding of the potential deforestation patterns in the undeveloped region under different scenarios of forest law changes.
This project is the culmination of our hard work and dedication over the past months. We would like to thank all contributors and our mentors for their support and guidance throughout this journey. While this marks the end of active development for this project, we encourage users and developers to use our work as a foundation for further research and development in this field.
Please refer to the project documentation for more detailed information about the project and how to use it