GIS and Machine Learning for Small Area Classifications in Developing Countries

Adegbola Ojo

• Publisher: CRC Press
• Year: 2020
• Print ISBN: 9780367322441
• eBook ISBN: 9781000289398
• Edition: 1
• Topic: Physical Sciences
• Subtopic: City Planning & Urban Development

Since the emergence of contemporary area classifications, population geography has witnessed a renaissance in the area of policy related spatial analysis. Area classifications subsume geodemographic systems which often use data mining techniques and machine learning algorithms to simplify large and complex bodies of information about people and the places in which they live, work and undertake other social activities. Outputs developed from the grouping of small geographical areas on the basis of multi- dimensional data have proved beneficial particularly for decision-making in the commercial sectors of a vast number of countries in the northern hemisphere. This book argues that small area classifications offer countries in the Global South a distinct opportunity to address human population policy related challenges in novel ways using area-based initiatives and evidence-based methods.

This book exposes researchers, practitioners, and students to small area segmentation techniques for understanding, interpreting, and visualizing the configuration, dynamics, and correlates of development policy challenges at small spatial scales. It presents strategic and operational responses to these challenges in cost effective ways. Using two developing countries as case studies, the book connects new transdisciplinary ways of thinking about social and spatial inequalities from a scientific perspective with GIS and Data Science. This offers all stakeholders a framework for engaging in practical dialogue on development policy within urban and rural settings, based on real-world examples.

Features:

• 
  
  

• The first book to address the huge potential of small area segmentation for sustainable development, combining explanations of concepts, a range of techniques, and current applications.

• 
  
  

• Includes case studies focused on core challenges that confront developing countries and provides thorough analytical appraisal of issues that resonate with audiences from the Global South.

• 
  
  

• Combines GIS and machine learning methods for studying interrelated disciplines such as Demography, Urban Science, Sociology, Statistics, Sustainable Development and Public Policy.

• 
  
  

• Uses a multi-method approach and analytical techniques of primary and secondary data.

• 
  
  

• Embraces a balanced, chronological, and well sequenced presentation of information, which is very practical for readers.

Part 1

Background, Concepts, and Definitions

1

Introduction

1.1 Global South or Developing World?

Prior to discussing some of the contemporary demographic trends and shifts, it is necessary to engage in some conceptual debates. What do we call the parts of the world that this book covers? Do we refer to these parts of the world as the Global South or developing world? Numerous development scholars continue to grapple with this important conceptual question.

Although the term “Global South” emerged in the 1950s, its first notable use dates back to 1969. Carl Preston Oglesby, an American writer, academic, and political activist, is credited with coining the term within contemporary political context. Oglesby acted as the editor of a special issue of the liberal Catholic journal called Commonweal. The special issue focused on the Vietnam War and Oglesby (1969) argued that centuries of US dominance over the Global South have converged to produce an intolerable social order. The founding members of the Non-Aligned Movement (NAM) also used the term politically.¹ Since the late 1960s, the term “Global South” has amassed multiple nuances and meanings (Horner and Hulme, 2019). Some have described countries within the Global South as politically and culturally marginalized (Leimgruber, 2018).

The emergence of the concept stemmed from historical attempts to define economically poorer and richer parts of the world. Many scholars and practitioners consider its predecessors less favorable while acknowledging that the term “Global South” nevertheless remains imperfect. Mahler (2017) contends that the concept of Global South subsumes at least three notions. The first notion is that it refers to nation states that are economically disadvantaged. This interpretation stems from the use within the corridors of international organizations and development partners. Mahler further argues that a second notion of the concept of Global South exposes the externalities of capitalism. In a sense, the concept is largely used to differentiate those parts of the world that were subjugated by others especially during the colonial and indeed the post-colonial era. The third notion of the concept of Global South is an extension of the second one. Mahler (2017) suggests that some countries interpret the concept through the lens of a shared experience of subjugation triggered by global capitalism. Those countries identify their conditions as similar and they form movements that they believe can cater to their mutual interests.

Figure 1.1
Map of countries by World Bank Income Groups.

Questions remain as to the geographic limits of the Global South. Some scholars and practitioners generally use the phrase as a catchall for several countries located within four main geopolitical regions of the world: Africa, Asia, Latin America, and Oceania (Connell and Dados, 2014; Watson, 2016). For others, the literal geographic impression invoked by the term is that of the equator splitting the globe into northern and southern hemispheres. However, this is not necessarily the case because there are numerous countries within the northern hemisphere with human development and income characteristics that correspond to others within the southern hemisphere. Consequently, there is no agreed consensus in the literature regarding the designated geographic boundaries of the Global South. This works to its disadvantage. For the sake of consistency, some researchers and policymakers utilize the development and income characteristics illustrated in Figures 1.1 and 1.2, which are proposed by the World Bank and the United Nations, respectively.

Figure 1.2
Map of countries by United Nations Development Groups.

It is argued that countries in the world are too diverse to be compartmentalized into two boxes: one for south and one for north. Although the term “Global South” has been increasingly popularized, it is considered too vague and much less homogenizing. The alternative term “developing world” represents a radical move from the focus on cultural or developmental dissimilarities between countries. This term gives prominence to geopolitical power relations. There are arguments that the concept should be embraced to address those spaces and peoples that have been and are being routinely negatively impacted by contemporary capitalist globalization. The term “developing world” also brings about a connotation of improving conditions, the recognition that social and economic conditions are evolutionary processes. In the remainder of this book, the term “developing world” is broadly embraced because it is considered as a more homogenizing term which also admits that countries can advance their economic growth through better utilization of natural and human resources, and that this can further result in changes in social, political, and economic structures of nations. Although the World Bank hinted in 2016 that it is retiring the term “developing world,” the implication of this change is that it is likely to exacerbate global levels of inequality. There is still gross inequality within and between countries that needs to be addressed and the categorizations of developed and developing can help make the case for that.

1.2 Demographic Shifts across the Developing World

Drawing upon recent data released by the UN Population Division for the World Population Prospects 2019, we discovered that during the 70-year period between 1950 and 2020, the average annual global demographic growth stood at 1.60%. During this period, the demographic growth has been most pronounced within the least developed countries, which averages 2.41% annually. Similarly, low-income countries have witnessed the largest average annual population growth (2.48%). Table 1.1 summarizes the average annual change in population across six continents. Estimates shown in the table from 2020 and beyond are based on a medium fertility variant calculation. Focusing on the four continents that make up the developing world, one can discover some interesting trends. During the 150-year period, population growth will remain positive across Africa although the growth will diminish significantly to below 1% per annum from 2080 onward.

Table 1.1
Average annual rate of population change, 1950–2100 (percentage)

Period	Africa	Asia	Europe	Latin America & the Caribbean	North America	Oceania
1950 to 1955	2.08	1.95	0.97	2.65	1.65	2.07
1955 to 1960	2.29	1.92	0.97	2.69	1.76	2.14
1960 to 1965	2.44	2.11	0.95	2.71	1.40	2.07
1965 to 1970	2.54	2.46	0.69	2.54	1.03	2.29
1970 to 1975	2.64	2.28	0.60	2.37	0.95	1.72
1975 to 1980	2.78	1.97	0.49	2.25	0.95	1.34
1980 to 1985	2.82	1.95	0.40	2.14	0.95	1.61
1985 to 1990	2.78	1.99	0.37	1.93	0.98	1.64
1990 to 1995	2.58	1.59	0.17	1.74	1.02	1.48
1995 to 2000	2.46	1.37	–0.04	1.55	1.19	1.34
2000 to 2005	2.44	1.23	0.10	1.32	0.93	1.39
2005 to 2010	2.52	1.13	0.19	1.18	0.96	1.81
2010 to 2015	2.58	1.04	0.18	1.07	0.79	1.56
2015 to 2020	2.51	0.92	0.12	0.94	0.65	1.37
2020 to 2025	2.37	0.77	–0.05	0.84	0.59	1.21
2025 to 2030	2.25	0.62	–0.12	0.70	0.56	1.11
2030 to 2035	2.13	0.49	–0.17	0.56	0.53	1.02
2035 to 2040	2.01	0.36	–0.20	0.43	0.45	0.93
2040 to 2045	1.88	0.25	–0.23	0.32	0.38	0.86
2045 to 2050	1.74	0.14	–0.26	0.22	0.34	0.80
2050 to 2055	1.61	0.04	–0.29	0.11	0.32	0.74
2055 to 2060	1.48	–0.05	–0.33	0.02	0.33	0.69
2060 to 2065	1.36	–0.12	–0.34	–0.07	0.34	0.63
2065 to 2070	1.24	–0.19	–0.32	–0.16	0.33	0.59
2070 to 2075	1.12	–0.25	–0.28	–0.24	0.30	0.54
2075 to 2080	1.01	–0.29	–0.24	–0.30	0.27	0.50
2080 to 2085	0.90	–0.33	–0.19	–0.36	0.25	0.46
2085 to 2090	0.80	–0.35	–0.16	–0.40	0.24	0.43
2090 to 2095	0.70	–0.37	–0.14	–0.44	0.24	0.40
2095 to 2100	0.61	–0.39	–0.14	–0.46	0.25	0.37

Source: Author’s elaborations based on data from the United Nations, Department of Economic and Social Affairs, Population Division (2019). World Population Prospects 2019, Online Edition.

Oceania will also experience a positive population growth during this period for which figures are shown. However, the growth will diminish below 1% per annum sooner than in Africa from 2035 onward. Unlike Africa and Oceania where growth will remain positive, the populations of Asia and Latin America and the Caribbean are expected to have a decline at some point. The populations of Asia will begin to decline by 2055, while that of Latin America and the Caribbean will turn negative half a decade later – 2060.

Another story that can be told from the data presented in Table 1.1 is that the population of much of the developing world is expected to rise faster than other parts of the world especially during the 21st century. Africa will mostly contribute to this demographic trend and by the middle of the century, it is estimated that the continent would have doubled its population to about 2.5 billion people (PRB, 2016).

Different factors contribute toward the population growth. However, various studies have agreed that there are two principal reasons for the rapid population growth in much of the developing world. The first major factor accounting for the population growth across the developing world is the sharp decline in the mortality rates of infants and children (Golding et al., 2017). Considerable progress has been made in reducing under-5 mortality especially in the last 30 years. Since 1990, it is estimated that the world has reduced under-5 mortality by around 58% (Hug et al., 2018).

The second important factor that has continued to contribute to population growth in developing countries is the limited reduction in birth rates. From a global point of view, fertility levels have reached unprecedented low levels. However, stark differences persist when patterns are observed ...

Table of contents

• Cover
• Half Title
• Title Page
• Copyright Page
• Dedication
• Contents
• Foreword
• Preface
• Acknowledgements
• Author Biography
• Abbreviations
• Part 1 Background, Concepts, and Definitions
• Part 2 Underlying Techniques and Deployment Approaches
• Part 3 Illustrative Applications and Conclusion
• Index