Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery

Abstract Forty million people are estimated to be in some form of modern slavery across the globe. Understanding the factors that make any particular individual or geographical region vulnerable to such abuse is essential for the development of effective interventions and policy. Efforts to isolate...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Rosa Lavelle-Hill, Gavin Smith, Anjali Mazumder, Todd Landman, James Goulding
Formato:	article
Lenguaje:	EN
Publicado:	Springer Nature 2021
Materias:	History of scholarship and learning. The humanities AZ20-999 Social Sciences H
Acceso en línea:	https://doaj.org/article/0bbe6d85f73e487abc3924b5c27bad30
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:0bbe6d85f73e487abc3924b5c27bad30
record_format	dspace
spelling	oai:doaj.org-article:0bbe6d85f73e487abc3924b5c27bad302021-11-21T12:28:27ZMachine learning methods for “wicked” problems: exploring the complex drivers of modern slavery10.1057/s41599-021-00938-z2662-9992https://doaj.org/article/0bbe6d85f73e487abc3924b5c27bad302021-11-01T00:00:00Zhttps://doi.org/10.1057/s41599-021-00938-zhttps://doaj.org/toc/2662-9992Abstract Forty million people are estimated to be in some form of modern slavery across the globe. Understanding the factors that make any particular individual or geographical region vulnerable to such abuse is essential for the development of effective interventions and policy. Efforts to isolate and assess the importance of individual drivers statistically are impeded by two key challenges: data scarcity and high dimensionality, typical of many “wicked problems”. The hidden nature of modern slavery restricts available data points; and the large number of candidate variables that are potentially predictive of slavery inflate the feature space exponentially. The result is a “small n, large p” setting, where overfitting and significant inter-correlation of explanatory variables can render more traditional statistical approaches problematic. Recent advances in non-parametric computational methods, however, offer scope to overcome such challenges and better capture the complex nature of modern slavery. We present an approach that combines non-linear machine-learning models and strict cross-validation methods with novel variable importance techniques, emphasising the importance of stability of model explanations via a Rashomon-set analysis. This approach is used to model the prevalence of slavery in 48 countries, with results bringing to light the importance of new predictive factors—such as a country’s capacity to protect the physical security of women, which has been previously under-emphasised in quantitative models. Further analyses uncover that women are particularly vulnerable to exploitation in areas where there is poor access to resources. Our model was then leveraged to produce new out-of-sample estimates of slavery prevalence for countries where no survey data currently exists.Rosa Lavelle-HillGavin SmithAnjali MazumderTodd LandmanJames GouldingSpringer NaturearticleHistory of scholarship and learning. The humanitiesAZ20-999Social SciencesHENHumanities & Social Sciences Communications, Vol 8, Iss 1, Pp 1-11 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	History of scholarship and learning. The humanities AZ20-999 Social Sciences H
spellingShingle	History of scholarship and learning. The humanities AZ20-999 Social Sciences H Rosa Lavelle-Hill Gavin Smith Anjali Mazumder Todd Landman James Goulding Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
description	Abstract Forty million people are estimated to be in some form of modern slavery across the globe. Understanding the factors that make any particular individual or geographical region vulnerable to such abuse is essential for the development of effective interventions and policy. Efforts to isolate and assess the importance of individual drivers statistically are impeded by two key challenges: data scarcity and high dimensionality, typical of many “wicked problems”. The hidden nature of modern slavery restricts available data points; and the large number of candidate variables that are potentially predictive of slavery inflate the feature space exponentially. The result is a “small n, large p” setting, where overfitting and significant inter-correlation of explanatory variables can render more traditional statistical approaches problematic. Recent advances in non-parametric computational methods, however, offer scope to overcome such challenges and better capture the complex nature of modern slavery. We present an approach that combines non-linear machine-learning models and strict cross-validation methods with novel variable importance techniques, emphasising the importance of stability of model explanations via a Rashomon-set analysis. This approach is used to model the prevalence of slavery in 48 countries, with results bringing to light the importance of new predictive factors—such as a country’s capacity to protect the physical security of women, which has been previously under-emphasised in quantitative models. Further analyses uncover that women are particularly vulnerable to exploitation in areas where there is poor access to resources. Our model was then leveraged to produce new out-of-sample estimates of slavery prevalence for countries where no survey data currently exists.
format	article
author	Rosa Lavelle-Hill Gavin Smith Anjali Mazumder Todd Landman James Goulding
author_facet	Rosa Lavelle-Hill Gavin Smith Anjali Mazumder Todd Landman James Goulding
author_sort	Rosa Lavelle-Hill
title	Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
title_short	Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
title_full	Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
title_fullStr	Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
title_full_unstemmed	Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
title_sort	machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery
publisher	Springer Nature
publishDate	2021
url	https://doaj.org/article/0bbe6d85f73e487abc3924b5c27bad30
work_keys_str_mv	AT rosalavellehill machinelearningmethodsforwickedproblemsexploringthecomplexdriversofmodernslavery AT gavinsmith machinelearningmethodsforwickedproblemsexploringthecomplexdriversofmodernslavery AT anjalimazumder machinelearningmethodsforwickedproblemsexploringthecomplexdriversofmodernslavery AT toddlandman machinelearningmethodsforwickedproblemsexploringthecomplexdriversofmodernslavery AT jamesgoulding machinelearningmethodsforwickedproblemsexploringthecomplexdriversofmodernslavery
_version_	1718418985349808128

Machine learning methods for “wicked” problems: exploring the complex drivers of modern slavery

Ejemplares similares