Probing factors associated with ecological footprint through machine learning

Sustainability is achieved by balancing economic, environmental, and social considerations, ensuring that present needs are met without depleting resources or causing irreparable harm to future generations. This is aided by the Ecological Footprint (EF) matrices that quantify the environmental impac...

Full description

Saved in:
Bibliographic Details
Main Authors: Chupeco, Arabelle Raisa C., Recel, Cher Danica T., Martinez, John Albert R.
Format: text
Language:English
Published: Animo Repository 2023
Subjects:
Online Access:https://animorepository.dlsu.edu.ph/etdb_bio/72
https://animorepository.dlsu.edu.ph/context/etdb_bio/article/1075/viewcontent/2024_Chupeco_EtAl_Probing_Factors_Associated_with_Ecological_Footprint_Through_Mach.pdf
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: De La Salle University
Language: English
id oai:animorepository.dlsu.edu.ph:etdb_bio-1075
record_format eprints
spelling oai:animorepository.dlsu.edu.ph:etdb_bio-10752024-08-20T02:00:36Z Probing factors associated with ecological footprint through machine learning Chupeco, Arabelle Raisa C. Recel, Cher Danica T. Martinez, John Albert R. Sustainability is achieved by balancing economic, environmental, and social considerations, ensuring that present needs are met without depleting resources or causing irreparable harm to future generations. This is aided by the Ecological Footprint (EF) matrices that quantify the environmental impact by measuring resource consumption relative to the Earth's regenerative capacity. Policymakers, therefore, face the challenge of aligning these complex dynamics, driven by the pressing need to explore comprehensive indicators at the country level. To address this concern, this study analyzed the association between various country-level metrics and the ecological state of nations, drawing data respectively from World Bank and National Footprint Accounts for the year 2018. Logistic regression models were developed to identify which World Development Indicators are associated with EF based on country-level data. Machine learning techniques were employed in the R programming language to develop logistic regression models. These models were established for each variable class within specific data themes, aiming to predict a country's ecological state, i.e., whether it is relatively reserved or deficit. A strict benchmark required models to reach an Area Under the Curve of 60% or higher to be deemed acceptable. Additionally, the McFadden Pseudo R2 value, confined within 0.2 to 0.4, was employed to gauge the goodness of fit for each model. This approach yielded eight models, with 15 country-level predictors as statistically significant. Deficit countries were found to be associated with increased mortality due to road traffic injuries, access to electricity, basic sanitation services, safe drinking water services, GFP per person employed, value added by the industry and services (% GDP), and imports of goods and services. Conversely, countries that are relatively ecologically reserved were associated with higher crude birth rate, crude death rate, under-5 mortality rate, adolescent fertility rate, renewable energy consumption, industry value added per worker, and poverty gap. The associations identified in this study facilitate a deeper understanding of real-world factors that influence sustainability, thus, enabling environmental, economic, and societal interventions to be formulated and deployed. 2023-11-01T07:00:00Z text application/pdf https://animorepository.dlsu.edu.ph/etdb_bio/72 https://animorepository.dlsu.edu.ph/context/etdb_bio/article/1075/viewcontent/2024_Chupeco_EtAl_Probing_Factors_Associated_with_Ecological_Footprint_Through_Mach.pdf Biology Bachelor's Theses English Animo Repository Nature—Effect of human beings on Machine learning Logistic regression analysis Biology
institution De La Salle University
building De La Salle University Library
continent Asia
country Philippines
Philippines
content_provider De La Salle University Library
collection DLSU Institutional Repository
language English
topic Nature—Effect of human beings on
Machine learning
Logistic regression analysis
Biology
spellingShingle Nature—Effect of human beings on
Machine learning
Logistic regression analysis
Biology
Chupeco, Arabelle Raisa C.
Recel, Cher Danica T.
Martinez, John Albert R.
Probing factors associated with ecological footprint through machine learning
description Sustainability is achieved by balancing economic, environmental, and social considerations, ensuring that present needs are met without depleting resources or causing irreparable harm to future generations. This is aided by the Ecological Footprint (EF) matrices that quantify the environmental impact by measuring resource consumption relative to the Earth's regenerative capacity. Policymakers, therefore, face the challenge of aligning these complex dynamics, driven by the pressing need to explore comprehensive indicators at the country level. To address this concern, this study analyzed the association between various country-level metrics and the ecological state of nations, drawing data respectively from World Bank and National Footprint Accounts for the year 2018. Logistic regression models were developed to identify which World Development Indicators are associated with EF based on country-level data. Machine learning techniques were employed in the R programming language to develop logistic regression models. These models were established for each variable class within specific data themes, aiming to predict a country's ecological state, i.e., whether it is relatively reserved or deficit. A strict benchmark required models to reach an Area Under the Curve of 60% or higher to be deemed acceptable. Additionally, the McFadden Pseudo R2 value, confined within 0.2 to 0.4, was employed to gauge the goodness of fit for each model. This approach yielded eight models, with 15 country-level predictors as statistically significant. Deficit countries were found to be associated with increased mortality due to road traffic injuries, access to electricity, basic sanitation services, safe drinking water services, GFP per person employed, value added by the industry and services (% GDP), and imports of goods and services. Conversely, countries that are relatively ecologically reserved were associated with higher crude birth rate, crude death rate, under-5 mortality rate, adolescent fertility rate, renewable energy consumption, industry value added per worker, and poverty gap. The associations identified in this study facilitate a deeper understanding of real-world factors that influence sustainability, thus, enabling environmental, economic, and societal interventions to be formulated and deployed.
format text
author Chupeco, Arabelle Raisa C.
Recel, Cher Danica T.
Martinez, John Albert R.
author_facet Chupeco, Arabelle Raisa C.
Recel, Cher Danica T.
Martinez, John Albert R.
author_sort Chupeco, Arabelle Raisa C.
title Probing factors associated with ecological footprint through machine learning
title_short Probing factors associated with ecological footprint through machine learning
title_full Probing factors associated with ecological footprint through machine learning
title_fullStr Probing factors associated with ecological footprint through machine learning
title_full_unstemmed Probing factors associated with ecological footprint through machine learning
title_sort probing factors associated with ecological footprint through machine learning
publisher Animo Repository
publishDate 2023
url https://animorepository.dlsu.edu.ph/etdb_bio/72
https://animorepository.dlsu.edu.ph/context/etdb_bio/article/1075/viewcontent/2024_Chupeco_EtAl_Probing_Factors_Associated_with_Ecological_Footprint_Through_Mach.pdf
_version_ 1808617037908410368