IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING

<p align="justify">Climate change and biodiversity loss are major challenges that could lead to something like mass extinction after 66 million years. Currently, over 41,000 species are threatened with extinction, accounting for 28% of the world's species. This is due to rapid u...

Full description

Saved in:
Bibliographic Details
Main Author: Santoso, Cokro
Format: Final Project
Language:Indonesia
Online Access:https://digilib.itb.ac.id/gdl/view/73425
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Institut Teknologi Bandung
Language: Indonesia
id id-itb.:73425
spelling id-itb.:734252023-06-20T10:24:46ZIDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING Santoso, Cokro Indonesia Final Project Biodiversity, Machine Learning, Remote Sensing INSTITUT TEKNOLOGI BANDUNG https://digilib.itb.ac.id/gdl/view/73425 <p align="justify">Climate change and biodiversity loss are major challenges that could lead to something like mass extinction after 66 million years. Currently, over 41,000 species are threatened with extinction, accounting for 28% of the world's species. This is due to rapid urbanization and development projects, disasters, fires, and conflicts between animals and humans. The ecological and socio-economic impacts of human interactions that destroy crops and extreme weather events such as droughts, floods, and landslides that cause shifts in native habitats. Attacks carried out by animals in areas around settlements that have food such as rice fields, plantations, etc. Thus, further research is needed on the potential conflict zones between humans and animals using global remote sensing data using machine learning and the impact of global environmental-climate pressures and shifts based on food sources. The results obtained Global Mammal Habitat suitability index with Random forest, SVM, MaxEnt, and Agreement accuracy are 0.83 (High), 0.721 (Medium), 0.801 (High), and 0.800 (High), respectively. The Environmental Stress Index that has been created is the Standardarized Precipitation Index (SPI) to identify drought has a level of significance based on La Nina and El Nino, and Potential Conflict Locations in 4 month periods there are several locations that have the highest potential for conflict including North Africa, South Africa, and Central America. text
institution Institut Teknologi Bandung
building Institut Teknologi Bandung Library
continent Asia
country Indonesia
Indonesia
content_provider Institut Teknologi Bandung
collection Digital ITB
language Indonesia
description <p align="justify">Climate change and biodiversity loss are major challenges that could lead to something like mass extinction after 66 million years. Currently, over 41,000 species are threatened with extinction, accounting for 28% of the world's species. This is due to rapid urbanization and development projects, disasters, fires, and conflicts between animals and humans. The ecological and socio-economic impacts of human interactions that destroy crops and extreme weather events such as droughts, floods, and landslides that cause shifts in native habitats. Attacks carried out by animals in areas around settlements that have food such as rice fields, plantations, etc. Thus, further research is needed on the potential conflict zones between humans and animals using global remote sensing data using machine learning and the impact of global environmental-climate pressures and shifts based on food sources. The results obtained Global Mammal Habitat suitability index with Random forest, SVM, MaxEnt, and Agreement accuracy are 0.83 (High), 0.721 (Medium), 0.801 (High), and 0.800 (High), respectively. The Environmental Stress Index that has been created is the Standardarized Precipitation Index (SPI) to identify drought has a level of significance based on La Nina and El Nino, and Potential Conflict Locations in 4 month periods there are several locations that have the highest potential for conflict including North Africa, South Africa, and Central America.
format Final Project
author Santoso, Cokro
spellingShingle Santoso, Cokro
IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING
author_facet Santoso, Cokro
author_sort Santoso, Cokro
title IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING
title_short IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING
title_full IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING
title_fullStr IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING
title_full_unstemmed IDENTIFICATION OF GLOBAL HUMAN-WILDLIFE CONFLICT POTENTIAL ZONES BASED ON MULTI-SOURCE REMOTE SENSING DATA AND MACHINE LEARNING
title_sort identification of global human-wildlife conflict potential zones based on multi-source remote sensing data and machine learning
url https://digilib.itb.ac.id/gdl/view/73425
_version_ 1822279587198926848