scholarly journals Integrating active learning and crowdsourcing into large-scale supervised landcover mapping algorithms

2017 ◽  
Author(s):  
Stephanie R Debats ◽  
Lyndon D Estes ◽  
David R Thompson ◽  
Kelly K Caylor
Keyword(s):  
Active Learning ◽  
Large Scale ◽  
Learning Algorithm ◽  
Training Data ◽  
Sub Saharan Africa ◽  
Data Sets ◽  
Field Patterns ◽  
Sub Saharan ◽  
Highly Correlated ◽  
Computational Resources

Sub-Saharan Africa and other developing regions of the world are dominated by smallholder farms, which are characterized by small, heterogeneous, and often indistinct field patterns. In previous work, we developed an algorithm for mapping both smallholder and commercial agricultural fields that includes efficient extraction of a vast set of simple, highly correlated, and interdependent features, followed by a random forest classifier. In this paper, we demonstrated how active learning can be incorporated in the algorithm to create smaller, more efficient training data sets, which reduced computational resources, minimized the need for humans to hand-label data, and boosted performance. We designed a patch-based uncertainty metric to drive the active learning framework, based on the regular grid of a crowdsourcing platform, and demonstrated how subject matter experts can be replaced with fleets of crowdsourcing workers. Our active learning algorithm achieved similar performance as an algorithm trained with randomly selected data, but with 62% less data samples.

scholarly journals Integrating active learning and crowdsourcing into large-scale supervised landcover mapping algorithms

2017 ◽  
Author(s):  
Stephanie R Debats ◽  
Lyndon D Estes ◽  
David R Thompson ◽  
Kelly K Caylor
Keyword(s):  
Active Learning ◽  
Large Scale ◽  
Learning Algorithm ◽  
Training Data ◽  
Sub Saharan Africa ◽  
Data Sets ◽  
Field Patterns ◽  
Sub Saharan ◽  
Highly Correlated ◽  
Computational Resources

Sub-Saharan Africa and other developing regions of the world are dominated by smallholder farms, which are characterized by small, heterogeneous, and often indistinct field patterns. In previous work, we developed an algorithm for mapping both smallholder and commercial agricultural fields that includes efficient extraction of a vast set of simple, highly correlated, and interdependent features, followed by a random forest classifier. In this paper, we demonstrated how active learning can be incorporated in the algorithm to create smaller, more efficient training data sets, which reduced computational resources, minimized the need for humans to hand-label data, and boosted performance. We designed a patch-based uncertainty metric to drive the active learning framework, based on the regular grid of a crowdsourcing platform, and demonstrated how subject matter experts can be replaced with fleets of crowdsourcing workers. Our active learning algorithm achieved similar performance as an algorithm trained with randomly selected data, but with 62% less data samples.

State Failure and Challenges to Democratization in Africa

2009 ◽  
Vol 7 (2) ◽  
pp. 359-361 ◽  
Author(s):  
Jeffrey Herbst
Keyword(s):  
Large Scale ◽  
Culture Wars ◽  
Sub Saharan Africa ◽  
Data Sets ◽  
State Institutions ◽  
Different Types ◽  
Important Approach ◽  
Game Theoretic ◽  
Sub Saharan ◽  
Cross National

Robert H. Bates's new book When Things Fell Apart seeks to make a contribution in two different areas. Explicitly, it joins a large literature on why state institutions collapsed in sub-Saharan Africa, especially why leaders drove one economy after the next into the ground. Less emphatically stated but clear enough from the book's content and its structure is an important contribution to political science's “culture wars” over the use of different types of evidence, especially the sometimes competing claims for the primacy of country knowledge, game-theoretic modeling, and large cross-national data sets. In particular, Bates uses a deductive approach, where game-theoretic approaches are married to national outcomes through a deep immersion in the literature and intuition (a concept he clearly seeks to rehabilitate) and then tested by the use of a significant and original database that is nonetheless relegated to an appendix. This is a particularly important approach because no one would accuse Bates of being at all hostile to large-scale quantitative analysis. Indeed, the significant data collection at Harvard's Africa Research Program (http://africa.gov.harvard.edu), which he helped found, is a service to the discipline.

scholarly journals A generalized computer vision approach to mapping crop fields in heterogeneous agricultural landscapes

2015 ◽  
Author(s):  
Stephanie Debats ◽  
Dee Luo ◽  
Lyndon Estes ◽  
Thomas J Fuchs ◽  
Kelly K Caylor
Keyword(s):  
High Performance ◽  
Agricultural Land ◽  
Agricultural Landscapes ◽  
Sub Saharan Africa ◽  
Sensitivity Analyses ◽  
Crop Fields ◽  
Field Patterns ◽  
Multi Temporal ◽  
Sub Saharan ◽  
Highly Correlated

Smallholder farms dominate in many parts of the world, particularly Sub-Saharan Africa. These systems are characterized by small, heterogeneous, and often indistinct field patterns, requiring a specialized methodology to map agricultural land cover. Using a variety of sites in South Africa, we present a new approach to mapping agricultural fields, based on efficient extraction of a vast set of simple, highly correlated, and interdependent features, followed by a random forest classifier. We achieved similar high performance across agricultural types, including the spectrally indistinct smallholder fields as well as the more easily distinguishable commercial fields, and demonstrated the ability to generalize performance across large geographic areas. In sensitivity analyses, we determined multi-temporal information provided greater gains in accuracy than multi-spectral information.

scholarly journals A generalized computer vision approach to mapping crop fields in heterogeneous agricultural landscapes

2015 ◽  
Author(s):  
Stephanie Debats ◽  
Dee Luo ◽  
Lyndon Estes ◽  
Thomas J Fuchs ◽  
Kelly K Caylor
Keyword(s):  
High Performance ◽  
Agricultural Land ◽  
Agricultural Landscapes ◽  
Sub Saharan Africa ◽  
Sensitivity Analyses ◽  
Crop Fields ◽  
Field Patterns ◽  
Multi Temporal ◽  
Sub Saharan ◽  
Highly Correlated

Smallholder farms dominate in many parts of the world, particularly Sub-Saharan Africa. These systems are characterized by small, heterogeneous, and often indistinct field patterns, requiring a specialized methodology to map agricultural land cover. Using a variety of sites in South Africa, we present a new approach to mapping agricultural fields, based on efficient extraction of a vast set of simple, highly correlated, and interdependent features, followed by a random forest classifier. We achieved similar high performance across agricultural types, including the spectrally indistinct smallholder fields as well as the more easily distinguishable commercial fields, and demonstrated the ability to generalize performance across large geographic areas. In sensitivity analyses, we determined multi-temporal information provided greater gains in accuracy than multi-spectral information.

scholarly journals Precision Agriculture for Resource Use Efficiency in Smallholder Farming Systems in Sub-Saharan Africa: A Systematic Review

2021 ◽  
Vol 13 (3) ◽  
pp. 1158
Author(s):  
Cecilia M. Onyango ◽  
Justine M. Nyaga ◽  
Johanna Wetterlind ◽  
Mats Söderström ◽  
Kristin Piikki
Keyword(s):  
Resource Use ◽  
Precision Agriculture ◽  
Large Scale ◽  
English Language ◽  
Management Practices ◽  
Smallholder Farmers ◽  
Management Strategies ◽  
Sub Saharan Africa ◽  
Smallholder Farming ◽  
Sub Saharan

Opportunities exist for adoption of precision agriculture technologies in all parts of the world. The form of precision agriculture may vary from region to region depending on technologies available, knowledge levels and mindsets. The current review examined research articles in the English language on precision agriculture practices for increased productivity among smallholder farmers in Sub-Saharan Africa. A total of 7715 articles were retrieved and after screening 128 were reviewed. The results indicate that a number of precision agriculture technologies have been tested under SSA conditions and show promising results. The most promising precision agriculture technologies identified were the use of soil and plant sensors for nutrient and water management, as well as use of satellite imagery, GIS and crop-soil simulation models for site-specific management. These technologies have been shown to be crucial in attainment of appropriate management strategies in terms of efficiency and effectiveness of resource use in SSA. These technologies are important in supporting sustainable agricultural development. Most of these technologies are, however, at the experimental stage, with only South Africa having applied them mainly in large-scale commercial farms. It is concluded that increased precision in input and management practices among SSA smallholder farmers can significantly improve productivity even without extra use of inputs.

Governance and Regional Variation of Homicide Rates

2016 ◽  
Vol 61 (1) ◽  
pp. 25-45 ◽  
Author(s):  
Liqun Cao ◽  
Yan Zhang
Keyword(s):  
Spatial Regression ◽  
Sub Saharan Africa ◽  
Development Economic ◽  
Data Sets ◽  
Asian Countries ◽  
Ceteris Paribus ◽  
Region Data ◽  
Sub Saharan ◽  
Criminological Theories ◽  
Cross National

Criminological theories of cross-national studies of homicide have underestimated the effects of quality governance of liberal democracy and region. Data sets from several sources are combined and a comprehensive model of homicide is proposed. Results of the spatial regression model, which controls for the effect of spatial autocorrelation, show that quality governance, human development, economic inequality, and ethnic heterogeneity are statistically significant in predicting homicide. In addition, regions of Latin America and non-Muslim Sub-Saharan Africa have significantly higher rates of homicides ceteris paribus while the effects of East Asian countries and Islamic societies are not statistically significant. These findings are consistent with the expectation of the new modernization and regional theories.

Intraclass Correlation Coefficients for Designing Cluster-Randomized Trials in Sub-Saharan Africa Education

2016 ◽  
Vol 40 (6) ◽  
pp. 500-525 ◽  
Author(s):  
Ben Kelcey ◽  
Zuchao Shen ◽  
Jessaca Spybrook
Keyword(s):  
Large Scale ◽  
Randomized Trials ◽  
Intraclass Correlation ◽  
Educational Interventions ◽  
Sub Saharan Africa ◽  
Design Stage ◽  
Education Outcomes ◽  
Randomized Studies ◽  
Sub Saharan ◽  
Variance Explained

Objective: Over the past two decades, the lack of reliable empirical evidence concerning the effectiveness of educational interventions has motivated a new wave of research in education in sub-Saharan Africa (and across most of the world) that focuses on impact evaluation through rigorous research designs such as experiments. Often these experiments draw on the random assignment of entire clusters, such as schools, to accommodate the multilevel structure of schooling and the theory of action underlying many school-based interventions. Planning effective and efficient school randomized studies, however, requires plausible values of the intraclass correlation coefficient (ICC) and the variance explained by covariates during the design stage. The purpose of this study was to improve the planning of two-level school-randomized studies in sub-Saharan Africa by providing empirical estimates of the ICC and the variance explained by covariates for education outcomes in 15 countries. Method: Our investigation drew on large-scale representative samples of sixth-grade students in 15 countries in sub-Saharan Africa and includes over 60,000 students across 2,500 schools. We examined two core education outcomes: standardized achievement in reading and mathematics. We estimated a series of two-level hierarchical linear models with students nested within schools to inform the design of two-level school-randomized trials. Results: The analyses suggested that outcomes were substantially clustered within schools but that the magnitude of the clustering varied considerably across countries. Similarly, the results indicated that covariance adjustment generally reduced clustering but that the prognostic value of such adjustment varied across countries.

Land conflicts and social differentiation in eastern Uganda

2017 ◽  
Vol 55 (3) ◽  
pp. 395-422 ◽  
Author(s):  
Matt Kandel
Keyword(s):  
Large Scale ◽  
Local Level ◽  
Sub Saharan Africa ◽  
Social Differentiation ◽  
Land Conflicts ◽  
Current Period ◽  
Post Conflict ◽  
Sub Saharan ◽  
Historical Developments ◽  
Eastern Uganda

ABSTRACTRising competition and conflict over land in rural sub-Saharan Africa continues to attract the attention of researchers. Recent work has especially focused on land governance, post-conflict restructuring of tenure relations, and large-scale land acquisitions. A less researched topic as of late, though one deserving of greater consideration, pertains to how social differentiation on the local-level shapes relations to land, and how these processes are rooted in specific historical developments. Drawing on fieldwork conducted in Teso sub-region of eastern Uganda, this paper analyses three specific land conflicts and situates them within a broad historical trajectory. I show how each dispute illuminates changes in class relations in Teso since the early 1990s. I argue that this current period of socioeconomic transformation, which includes the formation of a more clearly defined sub-regional middle class and elite, constitutes the most prominent period of social differentiation in Teso since the early 20th century.

scholarly journals Resource frontiers and agglomeration economies: The varied logics of transnational land-based investing in Southern and Eastern Africa

AMBIO ◽  
2022 ◽  
Author(s):  
Dilini Abeygunawardane ◽  
Angela Kronenburg García ◽  
Zhanli Sun ◽  
Daniel Müller ◽  
Almeida Sitoe ◽  
...  
Keyword(s):  
Spatial Data ◽  
Large Scale ◽  
Agglomeration Economies ◽  
Sub Saharan Africa ◽  
Eastern Africa ◽  
Track Record ◽  
Level Data ◽  
Forestry Investments ◽  
Sub Saharan ◽  
Investment Location

AbstractActor-level data on large-scale commercial agriculture in Sub-Saharan Africa are scarce. The peculiar choice of transnational investing in African land has, therefore, been subject to conjecture. Addressing this gap, we reconstructed the underlying logics of investment location choices in a Bayesian network, using firm- and actor-level interview and spatial data from 37 transnational agriculture and forestry investments across 121 sites in Mozambique, Zambia, Tanzania, and Ethiopia. We distinguish four investment locations across gradients of resource frontiers and agglomeration economies to derive the preferred locations of different investors with varied skillsets and market reach (i.e., track record). In contrast to newcomers, investors with extensive track records are more likely to expand the land use frontier, but they are also likely to survive the high transaction costs of the pre-commercial frontier. We highlight key comparative advantages of Southern and Eastern African frontiers and map the most probable categories of investment locations.

Sign in / Sign up

Export Citation Format

Share Document