Sort by
Forecasting first-year student mobility using explainable machine learning techniques

In the context of regional sciences and migration studies, gravity and radiation models are typically used to estimate human spatial mobility of all kinds. These formal models are incorporated as part of regression models along with co-variates, to better represent regional specific aspects. Often, the correlations between dependent and independent variables are of non-linear type and follow complex spatial interactions and multicollinearity. To address some of the model-related obstacles and to arrive at better predictions, we introduce machine learning algorithm class XGBoost to the estimation of spatial interactions and provide useful statistics and visual representations for the model evaluation and the evaluation and interpretation of the independent variables. The methods suggested are used to study the case of the spatial mobility of high-school graduates to the enrolment in higher education institutions in Germany at the county-level. We show that machine learning techniques can deliver explainable results that compare to traditional regression modeling. In addition to typically high model fits, variable-based indicators such as the Shapley Additive Explanations value (SHAP) provide significant additional information on the differentiated and non-linear effect of the variable values. For instance, we provide evidence that the initial study location choice is not related to the quality of local labor-markets in general, as there are both, strong positive and strong negative effects of the local academic employment rates on the migration decision. When controlling for about 28 co-variates, the attractiveness of the study location itself is the most important single factor of influence, followed by the classical distance-related variables travel time (gravitation) and regional opportunities (radiation). We show that machine learning methods can be transparent, interpretable, and explainable, when employed with adequate domain-knowledge and flanked by additional calculations and visualizations related to the model evaluation.

Open Access
Relevant
Horizon Europe: a green window of opportunity for european peripheral regions?

AbstractAn emerging field of research suggests that the policy and societal pressures for a green transition represent a “green window of opportunity” for peripheral regions. These regions often lag behind in overall innovation performance and may suffer from being places that don’t matter. At the same time, these are exactly the regions that the European Union is trying to support through several programmes, including Horizon Europe. This paper investigates the participation of organisations from peripheral regions in environmental projects funded by the Horizon Europe programme. To account for the multidimensional nature of regional peripherality, we define peripheral regions from a geographical, innovation and socio-economic perspective. We then analyse the relationship between these dimensions of regional peripherality and the extent to which regions benefit from Horizon environmental innovation projects in terms of participation, amount of funding and position in the overall network of project consortia.Our findings show a greater participation in Horizon environmental innovation projects for regions in Southern and Northern Europe, while within-country peripherality is negatively related to participation. At the same time, regions that are lagging in terms of innovation and socio-economic performance also receive less of this specific funding. Overall, geographical peripherality only tells a part of the story as several “places that don’t matter” for innovation and economic dynamism are also unable to benefit from these specific green windows of opportunity.

Open Access
Relevant
The spatiotemporal socio-demography of the Tokyo capital region: a data-driven explorative approach

AbstractIn the coming decades, most of Asia’s population will reside in megacities, vast urban regions accommodating 10–30 million people. However, Asian megacities will be at the same time situated in the countries whose national population is projected to decline rapidly in the coming decades. Hence, for scholars and policymakers of Asian countries, understanding how the socio-demography of mature, post-growth, megacities will evolve within space and time is crucial to envision long-term and effective spatial governance. Prior studies have shown that varied migration patterns among socio-demographic groups lead to synchronized re-urbanization, post-suburbanization, and urban shrinkage in mature city regions. However, existing studies have limitations: they often exclude large Asian megacities, lack micro-scale analyses, and use predefined spatial typologies/divisions that obscure detailed patterns. To address these research gaps, this study investigated sub-municipal spatiotemporal patterns in Tokyo, the largest Asian megacity, using micro-scale job-household data and unsupervised machine learning clustering. The study revealed that Tokyo, like Euro-American cities, has experienced regional synchronization of (re)urbanization and (post)suburbanization within a complex landscape of shrinkage. However, the synchronized sub/urban growth is not uniform across localities within Tokyo. Complex migration flows seem to create disparities in demographic growth and decline, emphasizing the need for collaborative governance among localities within a megacity. The study contributes to a wider audience who are interested not only in the evolution of cities but also in an emerging application of machine learning to quantitative urban analyses.

Open Access
Relevant