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

Yin, Ganmin; Huang, Zhou; Bao, Yi; Wang, Han; Li, Linna; Ma, Xiaolei; Zhang, Yi

doi:10.1007/s10707-022-00467-0

Cited by 23 publications

(5 citation statements)

References 40 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Meanwhile, the features expressed in a single NTL radiation intensity value are limited because of saturation and blooming effects (Levin et al., 2020; Peled & Fishman, 2021). For most geographical phenomena, neighboring characteristics are essential for determining central targets (Xing et al., 2020; Yin et al., 2022). Therefore, we build the spatiotemporal features of NTL data using the following steps:…”

Section: Methods and Datamentioning

confidence: 99%

High‐resolution quantification of building stock using multi‐source remote sensing imagery and deep learning

Bao

Huang

Wang

et al. 2023

J of Industrial Ecology

Self Cite

View full text Add to dashboard Cite

In recent decades, urbanization has led to an increase in building material stock. The high‐resolution quantification of building stock is essential to understand the spatial concentration of materials, urban mining potential, and sustainable urban development. Current approaches rely excessively on statistics or survey data, both of which are unavailable for most cities, particularly in underdeveloped areas. This study proposes an end‐to‐end deep‐learning model based on multi‐source remote sensing data, enabling the reliable estimation of building stock. Ground‐detail features extracted from optical remote sensing (ORS) and spatiotemporal features extracted from nighttime light (NTL) data are fused and incorporated into the model to improve accuracy. We also compare the performance of our feature‐fusion model with that of an ORS‐only regression model and traditional NTL regression for Beijing. The proposed model yields the best building‐stock estimation, with a Spearman's rank correlation coefficient of 0.69, weighted root mean square error of 0.58, and total error in the test set below 14%. Using gradient‐weighted class activation mapping, we further investigate the relationship between ORS features and building‐stock estimation. Our model exhibits reliable predictive capability and illustrates the tremendous value of the physical environment for estimating building stock. This research illustrates the significant potential of ORS and deep learning for stock estimation. Large‐scale, long‐term building‐stock investigations could also benefit from the end‐to‐end predictability and the data availability of the model.

show abstract

Section: Methods and Datamentioning

confidence: 99%

High‐resolution quantification of building stock using multi‐source remote sensing imagery and deep learning

Bao

Huang

Wang

et al. 2023

J of Industrial Ecology

Self Cite

View full text Add to dashboard Cite

show abstract

“…Drawing inspiration from Tobler's first law of geography (Tobler 2004), a geo-contextual multitask embedding learner (GMEL) is introduced for predicting commuting flows, which encodes geographic contextual information using graph's attention networks (GATs) (Liu et al 2020). Subsequently, a hybrid model featuring a preprocessing-encoder-decoder framework is devised for commuting flow prediction, leveraging geographical semantics and regional proximity effects (Yin et al 2023).…”

Section: Literature Reviewmentioning

confidence: 99%

“…Various studies have extended and transformed GCN models to enhance their performance, particularly in the context of predicting trip generations and traffic flows (Yin et al 2021). Despite the success of GCN models, their application in common traffic prediction may not be entirely suitable for commuting forecasts (Yin et al 2023). The unique nature of commuting travel introduces distinctive socioeconomic properties and individual characteristics that play pivotal roles in the prediction process (Gao et al 2022).…”

Section: Introductionmentioning

confidence: 99%

Forecasting the commuting generation using metropolis-informed GCN and the topological commuter portrait

Chen,

Zhao,

Chen

2024

Transportation

View full text Add to dashboard Cite

Understanding commuter traffic in transportation networks is crucial for sustainable urban planning with commuting generation forecasts operating as a pivotal stage in commuter traffic modeling. Overcoming challenges posed by the intricacy of commuting networks and the uncertainty of commuter behaviors, we propose MetroGCN, a metropolis-informed graph convolutional network designed for commuting forecasts in metropolitan areas. MetroGCN introduces dimensions of metropolitan indicators to comprehensively construct commuting networks with diverse socioeconomic features. This model also innovatively embeds topological commuter portraits in spatial interaction through a multi-graph representation approach capturing the semantic spatial correlations based on individual characteristics. By incorporating graph convolution and temporal convolution with a spatial–temporal attention module, MetroGCN adeptly handles high-dimensional dependencies in large commuting networks. Quantitative experiments on the Shenzhen metropolitan area datasets validate the superior performance of MetroGCN compared to state-of-the-art methods. Notably, the results highlight the significance of commuter age and income in forecasting commuting generations. Statistical significance analysis further underscores the importance of anthropic indicators for commuting production forecasts and environmental indicators for commuting attraction forecasts. This research contributes to technical advancement and valuable insights into the critical factors influencing commuting generation forecasts.

show abstract

“…The advancement of spatio-temporal graph neural networks has given rise to the field of traffic prediction, and it has been successfully applied to several traffic-related tasks, such as demand prediction [19], flow prediction [20,21] and driver maneuver anticipation [22]. Generally, various ST graph implementations of traffic prediction can be classified into three groups based on how to process time series: CNN-based, RNNbased and transformer-based approaches.…”

Section: Spatial-temporal Graph For Traffic Predictionmentioning

confidence: 99%

Traffic Flow Prediction on Graph with Node-level Curriculum Scheduler

Zhang

Wang

Fan

et al. 2023

Preprint

View full text Add to dashboard Cite

Traffic flow prediction is a critical part of the intelligent transportation system , and many schemes have been proposed for accurate flow prediction based on traffic sensors. However, existing approaches mainly capture spatial-temporal (ST) dependencies from the traffic graph and train their models using randomly ordered data. This approach ignores the fact that the modeling difficulty of each sensor/node from the traffic graph can vary significantly due to its spatial dependencies and temporal trends, leading to unreliable and unstable predictions for each traffic sensor in real-world applications. These issues motivated us to design a more effective training process for sensors with different levels of difficulty. The ST modeling difficulty stems from two factors: (I) deviations between the normal component and abnormal component in the time sequences; and (II) deviations between the learned relation and the explicit relation of a predetermined graph in the spatial dependencies. In this case, we argue that a well-designed curriculum with an easy-to-difficult order can benefit the training of ST models. This approach draws inspiration from the human learning principle that models should first learn straightforward concepts before tackling more difficult ones. To implement this approach, this paper proposes a method to measure the ST difficulty at the node level of traffic graphs from both spatial and temporal aspects. Then, it conducts a curriculum during the ST model training procedure. Specifically, 1 based on the ST difficulty score, the model training starts from a subgraph whose nodes are considered ”easy” because of their relatively consistent spatial relationships and regular temporal patterns. After that, more difficult nodes are gradually added to the subgraph and participate in the following training stages. We have conducted extensive experiments on two public traffic datasets, which show the superiority of our proposed curriculum scheduler over multiple state-of-the-art ST models.

show abstract

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

Cited by 23 publications

References 40 publications

High‐resolution quantification of building stock using multi‐source remote sensing imagery and deep learning

High‐resolution quantification of building stock using multi‐source remote sensing imagery and deep learning

Forecasting the commuting generation using metropolis-informed GCN and the topological commuter portrait

Traffic Flow Prediction on Graph with Node-level Curriculum Scheduler

Contact Info

Product

Resources

About