Limits...
A geovisual analytic approach to understanding geo-social relationships in the international trade network.

Luo W, Yin P, Di Q, Hardisty F, MacEachren AM - PLoS ONE (2014)

Bottom Line: This research aims to address this challenge through the framework of geovisual analytics.In addition to demonstrating the potential of visual analytics to provide insight concerning complex geo-social relationships at a global scale, the research also addresses the challenge of validating insights derived through interactive geovisual analytics.We develop two indicators to quantify the observed patterns, and then use a Monte-Carlo approach to support the hypothesis developed above.

View Article: PubMed Central - PubMed

Affiliation: GeoVISTA Center, Department of Geography, Pennsylvania State University, University Park, Pennsylvania, United States of America.

ABSTRACT
The world has become a complex set of geo-social systems interconnected by networks, including transportation networks, telecommunications, and the internet. Understanding the interactions between spatial and social relationships within such geo-social systems is a challenge. This research aims to address this challenge through the framework of geovisual analytics. We present the GeoSocialApp which implements traditional network analysis methods in the context of explicitly spatial and social representations. We then apply it to an exploration of international trade networks in terms of the complex interactions between spatial and social relationships. This exploration using the GeoSocialApp helps us develop a two-part hypothesis: international trade network clusters with structural equivalence are strongly 'balkanized' (fragmented) according to the geography of trading partners, and the geographical distance weighted by population within each network cluster has a positive relationship with the development level of countries. In addition to demonstrating the potential of visual analytics to provide insight concerning complex geo-social relationships at a global scale, the research also addresses the challenge of validating insights derived through interactive geovisual analytics. We develop two indicators to quantify the observed patterns, and then use a Monte-Carlo approach to support the hypothesis developed above.

Show MeSH

Related in: MedlinePlus

Validation results as a function of number of runs (N).
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC3928244&req=5

pone-0088666-g007: Validation results as a function of number of runs (N).

Mentions: The first approach examines whether the number of runs in each Monte-Carlo validation influences the final results. If results are robust, validation results will converge as the number of runs increases. Figure 7 displays the results in which the number of runs (N) is 1,000, 2,000, 5,000 and 10,000. When N is small, such as 1,000, the results display some reasonable fluctuations. As the number of runs rises, those results are smoothed and finally converge (as shown by the turquoise line on each plot representing 10,000 runs).


A geovisual analytic approach to understanding geo-social relationships in the international trade network.

Luo W, Yin P, Di Q, Hardisty F, MacEachren AM - PLoS ONE (2014)

Validation results as a function of number of runs (N).
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC3928244&req=5

pone-0088666-g007: Validation results as a function of number of runs (N).
Mentions: The first approach examines whether the number of runs in each Monte-Carlo validation influences the final results. If results are robust, validation results will converge as the number of runs increases. Figure 7 displays the results in which the number of runs (N) is 1,000, 2,000, 5,000 and 10,000. When N is small, such as 1,000, the results display some reasonable fluctuations. As the number of runs rises, those results are smoothed and finally converge (as shown by the turquoise line on each plot representing 10,000 runs).

Bottom Line: This research aims to address this challenge through the framework of geovisual analytics.In addition to demonstrating the potential of visual analytics to provide insight concerning complex geo-social relationships at a global scale, the research also addresses the challenge of validating insights derived through interactive geovisual analytics.We develop two indicators to quantify the observed patterns, and then use a Monte-Carlo approach to support the hypothesis developed above.

View Article: PubMed Central - PubMed

Affiliation: GeoVISTA Center, Department of Geography, Pennsylvania State University, University Park, Pennsylvania, United States of America.

ABSTRACT
The world has become a complex set of geo-social systems interconnected by networks, including transportation networks, telecommunications, and the internet. Understanding the interactions between spatial and social relationships within such geo-social systems is a challenge. This research aims to address this challenge through the framework of geovisual analytics. We present the GeoSocialApp which implements traditional network analysis methods in the context of explicitly spatial and social representations. We then apply it to an exploration of international trade networks in terms of the complex interactions between spatial and social relationships. This exploration using the GeoSocialApp helps us develop a two-part hypothesis: international trade network clusters with structural equivalence are strongly 'balkanized' (fragmented) according to the geography of trading partners, and the geographical distance weighted by population within each network cluster has a positive relationship with the development level of countries. In addition to demonstrating the potential of visual analytics to provide insight concerning complex geo-social relationships at a global scale, the research also addresses the challenge of validating insights derived through interactive geovisual analytics. We develop two indicators to quantify the observed patterns, and then use a Monte-Carlo approach to support the hypothesis developed above.

Show MeSH
Related in: MedlinePlus