Soft Computing Techniques For Forest Fires Prediction
Name
Muhammad Uzair
Abstract
Forests provide fresh air and they are the source of livelihood to a huge number of people around the world. The forest fires affect the human lives in different ways such as economic, environment and health and these forest fires are usually caused by human negligence, and other environmental factors. The forest fire problem is divided into categories, i.e. fire risk, forest fire detection, and the prediction of burnt area. Developing a system to cater for solutions to this problem can help to use the resources in an efficient manner and can also save human lives and wildlife. In this thesis, some soft computing techniques are discussed, to predict the burnt area from forest fires. The considered soft computing techniques are ANFIS, ANN, SVM. In particular, a new variant of ANFIS is presented. Two publicly available datasets were used: one from Portugal and one from Canada. The techniques were applied separately on both datasets and results were collected.
Graduation Thesis language
English
Graduation Thesis type
Master - Computer Science
Supervisor(s)
Dr. Stefania Tomasiello
Defence year
2021