Reconstructing commuters network using machine learning and urban indicators

  title={Reconstructing commuters network using machine learning and urban indicators},
  author={Gabriel Spadon and Andr{\'e} Carlos Ponce de Leon Ferreira de Carvalho and Jos{\'e} F. Rodrigues and Luiz G. A. Alves},
  journal={Scientific Reports},
Human mobility has a significant impact on several layers of society, from infrastructural planning and economics to the spread of diseases and crime. Representing the system as a complex network, in which nodes are assigned to regions (e.g., a city) and links indicate the flow of people between two of them, physics-inspired models have been proposed to quantify the number of people migrating from one city to the other. Despite the advances made by these models, our ability to predict the… 

Learning Mobility Flows from Urban Features with Spatial Interaction Models and Neural Networks**To appear in the Proceedings of 2020 IEEE International Conference on Smart Computing (SMARTCOMP 2020)

This paper proposes three neural network architectures, including graph neural networks (GNN), and conducts a systematic comparison between the proposed methods and state-of-the-art spatial interaction models, their modifications, and machine learning approaches.

Learning Geo-Contextual Embeddings for Commuting Flow Prediction

Geo-contextual Multitask Embedding Learner (GMEL) is proposed, a model that captures the spatial correlations from geographic contextual information for commuting flow prediction and a gradient boosting machine is trained based on the learned embeddings to predict commuting flows.

Learning-Based Travel Prediction in Urban Road Network Resilience Optimization

Urban mobility is a key part of routine operations in cities, but is increasingly at risk due to floods. To mitigate these risks, urban planning and disaster management agencies must anticipate the

ConvGCN-RF: A hybrid learning model for commuting flow prediction considering geographical semantics and neighborhood effects

This paper proposes a ‘preprocessing-encoder-decoder’ hybrid learning model, which can make full use of geographic semantic information and spatial neighborhood effects, thereby significantly improving the prediction performance of commuting flow prediction.

Survey of Machine Learning Methods Applied to Urban Mobility

This work presents a survey on urban mobility based on passengers’ data and machine learning methods, focusing on four applications for urban mobility: public datasets, passenger localization, detection of the transport mode and pattern recognition and generation of mobility models.

From Cities to Series: Complex Networks and Deep Learning for Improved Spatial and Temporal Analytics

A new neural network architecture capable of modeling dynamic processes observed in spatial and temporal data with applications in epidemics propagation, weather forecasting, and patient monitoring in intensive care units, and a machine-learning methodology for analyzing and predicting links in the scope of human mobility between all the cities of Brazil.

Forecasting the evolution of fast-changing transportation networks using machine learning

This work uses machine learning approaches to predict edge removal on a monthly time scale and finds that models trained on data for a given month predict edge removals for the same month with high accuracy.

Commuting network effect on urban wealth scaling

This approach significantly refines the description of urban GDP and shows that incoming commuters contribute to wealth creation in urban areas and changes in urban GDP related to proportionate changes in population and incoming commuters depend on the initial values of these quantities.

Mathematical models to explain the origin of urban scaling laws: a synthetic review

This paper reviews the main mathematical models present in the literature that aim at explaining the origin and emergence of urban scaling, and proposes a general framework that includes all gravity models analyzed as a particular case.

An attractiveness-based model for human mobility in all spatial ranges

In the past decade, various aspects of human mobility, from individual to population levels in both spatial and time scales, have been studied. However, existing human mobility models still fail to



Predicting commuter flows in spatial networks using a radiation model based on temporal ranges

This work shows that traffic can efficiently and accurately be computed from a range-limited, network betweenness type calculation and captures the log-normal distribution of the traffic and attains a high Pearson correlation coefficient when compared with real traffic.

Uncovering the spatial structure of mobility networks

A versatile method is proposed, which extracts a coarse-grained signature of mobility networks, under the form of a 2 × 2 matrix that separates the flows into four categories, and allows the determination of categories of networks, and in the mobility case, the classification of cities according to their commuting structure.

Uncovering space-independent communities in spatial networks

This paper argues in this paper for a careful treatment of the constraints imposed by space on network topology and proposes a modularity function adapted to spatial networks for community detection and focuses on the problem of community detection.

Topological Street-Network Characterization Through Feature-Vector and Cluster Analysis

The results show how the joint of global features describes urban indicators that are deep-rooted in the network’s topology and how they reveal characteristics and similarities among sets of cities that are separated from each other.

Structure of Urban Movements: Polycentric Activity and Entangled Hierarchical Flows

The large scale, real-time ‘Oyster’ card database of individual person movements in the London subway is utilized in an unprecedented manner to reveal the structure and organization of the city.

A universal model for mobility and migration patterns

A stochastic process capturing local mobility decisions that helps to derive commuting and mobility fluxes that require as input only information on the population distribution is introduced, significantly improving the predictive accuracy of most of the phenomena affected by mobility and transport processes.

Scale-Adjusted Metrics for Predicting the Evolution of Urban Indicators and Quantifying the Performance of Cities

It is shown that this scale-adjusted metric provides a more appropriate/informative summary of the evolution of urban indicators and reveals patterns that do not appear in the Evolution of per capita values of indicators obtained from Brazilian cities, and is strongly correlated with their past values by a linear correspondence and that they also display crosscorrelations among themselves.

The worldwide air transportation network: Anomalous centrality, community structure, and cities' global roles

It is found that the worldwide air transportation network is a scale-free small-world network, and it is demonstrated that the most connected cities are not necessarily the most central, resulting in anomalous values of the centrality.

Spatial Networks

  • M. Barthelemy
  • Computer Science
    Encyclopedia of Social Network Analysis and Mining
  • 2014