Zobraziť v EDS

Human bias and CNNs’ superior insights in satellite based poverty mapping

Uložené v:
Podrobná bibliografia
Názov: Human bias and CNNs’ superior insights in satellite based poverty mapping
Autori: Sarmadi, Hamid, Wahab, Ibrahim, Hall, Ola, Rögnvaldsson, Thorsteinn, Ohlsson, Mattias
Prispievatelia: Lund University, Faculty of Social Sciences, Departments of Administrative, Economic and Social Sciences, Department of Human Geography, Lunds universitet, Samhällsvetenskapliga fakulteten, Samhällsvetenskapliga institutioner och centrumbildningar, Institutionen för kulturgeografi och ekonomisk geografi, Originator, Lund University, Faculty of Science, Centre for Environmental and Climate Science (CEC), Lunds universitet, Naturvetenskapliga fakulteten, Centrum för miljö- och klimatvetenskap (CEC), Originator, Lund University, Profile areas and other strong research environments, Strategic research areas (SRA), eSSENCE: The e-Science Collaboration, Lunds universitet, Profilområden och andra starka forskningsmiljöer, Strategiska forskningsområden (SFO), eSSENCE: The e-Science Collaboration, Originator, Lund University, Profile areas and other strong research environments, Lund University Profile areas, LU Profile Area: Natural and Artificial Cognition, Lunds universitet, Profilområden och andra starka forskningsmiljöer, Lunds universitets profilområden, LU profilområde: Naturlig och artificiell kognition, Originator
Zdroj: Scientific Reports. 14
Predmety: Social Sciences, Other Social Sciences, Samhällsvetenskap, Annan samhällsvetenskap, Natural Sciences, Computer and Information Sciences, Computer graphics and computer vision, Naturvetenskap, Data- och informationsvetenskap (Datateknik), Datorgrafik och datorseende
Popis: Satellite imagery is a potent tool for estimating human wealth and poverty, especially in regions lacking reliable data. This study compares a range of poverty estimation approaches from satellite images, spanning from expert-based to fully machine learning-based methodologies. Human experts ranked clusters from the Tanzania DHS survey using high-resolution satellite images. Then expert-defined features were utilized in a machine learning algorithm to estimate poverty. An explainability method was applied to assess the importance and interaction of these features in poverty prediction. Additionally, a convolutional neural network (CNN) was employed to estimate poverty from medium-resolution satellite images of the same locations. Our analysis indicates that increased human involvement in poverty estimation diminishes accuracy compared to machine learning involvement, exemplified with the case of Tanzania. Expert defined features exhibited significant overlap and poor interaction when used together in a classifier. Conversely, the CNN-based approach outperformed human experts, demonstrating superior predictive capability with medium-resolution images. These findings highlight the importance of leveraging machine learning explainability methods to identify predictive elements that may be overlooked by human experts. This study advocates for the integration of emerging technologies with traditional methodologies to optimize data collection and analysis of poverty and welfare.
Prístupová URL adresa: https://doi.org/10.1038/s41598-024-74150-9
Databáza: SwePub
Popis
Abstrakt:Satellite imagery is a potent tool for estimating human wealth and poverty, especially in regions lacking reliable data. This study compares a range of poverty estimation approaches from satellite images, spanning from expert-based to fully machine learning-based methodologies. Human experts ranked clusters from the Tanzania DHS survey using high-resolution satellite images. Then expert-defined features were utilized in a machine learning algorithm to estimate poverty. An explainability method was applied to assess the importance and interaction of these features in poverty prediction. Additionally, a convolutional neural network (CNN) was employed to estimate poverty from medium-resolution satellite images of the same locations. Our analysis indicates that increased human involvement in poverty estimation diminishes accuracy compared to machine learning involvement, exemplified with the case of Tanzania. Expert defined features exhibited significant overlap and poor interaction when used together in a classifier. Conversely, the CNN-based approach outperformed human experts, demonstrating superior predictive capability with medium-resolution images. These findings highlight the importance of leveraging machine learning explainability methods to identify predictive elements that may be overlooked by human experts. This study advocates for the integration of emerging technologies with traditional methodologies to optimize data collection and analysis of poverty and welfare.
ISSN:20452322
DOI:10.1038/s41598-024-74150-9