Attention Based Graph Bi-LSTM Networks for Traffic Forecasting

Predicting traffic conditions is tremendously challenging since every road is highly dependent on each other, both spatially and temporally. Recently, to capture this spatial and temporal dependency, specially designed architectures such as graph convolutional networks and temporal convolutional networks have been introduced. While there has been remarkable progress in traffic forecasting, we found that deep-learning-based traffic forecasting models still fail in certain patterns, mainly in event situations (e.g., rapid speed drops). Although it is commonly accepted that these failures are due to unpredictable noise, we found that these failures can be corrected by considering previous failures. Specifically, we observe autocorrelated errors in these failures, which indicates that some predictable information remains. In this study, to capture the correlation of errors, we introduce ResCAL, a residual estimation module for traffic forecasting, as a widely applicable add-on module to existing traffic forecasting models. Our ResCAL calibrates the prediction of the existing models in real time by estimating future errors using previous errors and graph signals. Extensive experiments on METR-LA and PEMS-BAY demonstrate that our ResCAL can correctly capture the correlation of errors and correct the failures of various traffic forecasting models in event situations.

In recent years, the rapid development of global artificial intelligence technology has become an important force in promoting the accelerated development of science and technology, and industry. From finance, medical care, and education to intelligent manufacturing and smart cities, artificial intelligence technology is penetrating into all walks of life, bringing convenience to people's lives. Through traffic flow forecasting, people can better plan their own travel routes and reduce the risk of traffic congestion. Accurate traffic forecast information can help to improve people's travel efficiency and improve urban service level, which is the core technology in the construction of intelligent transportation systems. The task of traffic flow forecasting aims at forecasting the future traffic flow trend according to the historical data information in the traffic network. The key challenge is how to model the temporal and spatial dependence in traffic networks. In the time domain, traffic flow has obvious periodicity characteristics, especially based on the daily periodicity and weekly periodicity characteristics of human social laws. We use the time series Transformer with predefined periodicity information to extract time dependence. In the spatial domain, there are similar traffic patterns in some areas on the macro level and the relationship between adjacent road sections on the micro level. We use the graph convolutional network based on the geographical adjacency matrix and the graph convolutional network based on the sequence semantic similarity matrix to extract the micro and macro spatial dependence respectively. We combine the two deeply and design a traffic flow forecasting model based on a graph convolutional network and Transformer to extract the deep temporal-spatial correlation. Finally, we experiment on several real-world traffic data sets to verify the effectiveness of the model.

Traffic forecasting is a challenging task due to complex spatial and temporal dependencies across sensor locations and time. Interest in solving this task has increased, but current research focuses on manually constructing neural network architectures without the aid of neural architecture search (NAS). In our work, we explore evolutionary neural architecture search (ENAS) by deploying a genetic algorithm (GA) to find optimal neural network architectures for predicting traffic conditions. The search space for the GA consists of arbitrary combinations of dilated convolutions and graph convolutions for modelling temporal and spatial dependencies respectively, limited in complexity only by technical constraints. Experimental results show that model architectures obtained via GA are able to match the current state-of-the-art on traffic prediction benchmarks.

Landslide displacement prediction has garnered significant recognition as a pivotal component in realizing successful early warnings and implementing effective control measures. This task remains challenging as landslide deformation involves not only temporal dependency within time series data but also spatial dependence across various regions within landslides. The present study proposes a landslide spatiotemporal displacement forecasting model by introducing attention-based deep learning algorithms based on spatiotemporal analysis. The Maximal Information Coefficient (MIC) approach is employed to quantify the spatial and temporal correlations within the daily data of Global Navigation Satellite System (GNSS) observations. Based on the quantitative spatiotemporal analysis, the proposed prediction model combines a convolutional neural network (CNN) and long short-term memory (LSTM) network to capture spatial and temporal dependencies individually. Spatial–temporal attention mechanisms are implemented to optimize the model. Additionally, we develop a single-point prediction model using LSTM and a multiple-point prediction model using the CNN-LSTM without an attention mechanism to compare the forecasting capabilities of the attention-based CNN-LSTM model. The Outang landslide in the Three Gorges Reservoir Area (TGRA), characterized by a large and active landslide equipped with an advanced monitoring system, is taken as a studied case. The temporal MIC results shed light on the response times of monitored daily displacement to external factors, showing a lagging duration of between 10 and 50 days. The spatial MIC results indicate mutual influence among different locations within the landslide, particularly in the case of nearby sites experiencing significant deformation. The attention-based CNN-LSTM model demonstrates an impressive predictive performance across six monitoring stations within the Outang landslide area. Notably, it achieves a remarkable maximum coefficient of determination (R2) value of 0.9989, accompanied by minimum values for root mean squared error (RMSE), absolute mean error (MAE), and mean absolute percentage error (MAPE), specifically, 1.18 mm, 0.99 mm, and 0.33%, respectively. The proposed model excels in predicting displacements at all six monitoring points, whereas other models demonstrate strong performance at specific individual stations but lack consistent performance across all stations. This study, involving quantitative deformation characteristics analysis and spatiotemporal displacement prediction, holds promising potential for a more profound understanding of landslide evolution and a significant contribution to reducing landslide risk.

Traffic flow forecasting is the basis for the dynamic control and application of Intelligent Transport Systems (ITS), which is of great practical importance in reducing road congestion. Accurately predicting traffic flow remains a significant obstacle due to the intricate spatial and temporal dependencies associated with traffic movement. To capture the simultaneous dynamics of traffic flow’s spatiotemporal features, this paper suggests a novel Spatiotemporal Interactive Dynamic Synchronous Graph Convolution Network (STIDSG) for traffic flow forecasting, which is composed of the Interactive Dynamic Graph Convolution Network (IDGCN) and the Spatiotemporal Synchronous Graph Convolution Layer (STSGCL). IDGCN learns and shares features using an interactive learning strategy that is based on Dynamic Graph Convolution (DGCN). By incorporating DGCN into the interactive learning structure, it is possible to learn dynamic spatial features of traffic flow in an interactive manner while also capturing temporal dependencies. STSGCL effectively captures the intricate local and dynamic correlations of traffic flows by designing numerous Spatiotemporal Synchronization Graph Convolution Modules (STSGCM) for different time periods. The results of the experiment demonstrate that the STIDSG model suggested in this paper can proficiently extract the changing spatiotemporal features of the traffic flow and outperforms the commonly used baseline methods for prediction. The results of the experiment demonstrate that the STIDSG model suggested in this paper can proficiently extract the changing spatiotemporal features of the traffic flow and outperforms the commonly used baseline methods for prediction.

This paper presents an innovative approach for enhancing power output forecasting of Photovoltaic (PV) power plants in dynamic environmental conditions using a Hybrid Deep Learning Model (DLM). The hybrid DLM employs a synergy of Convolutional Neural Network (CNN), Long Short-Term Memory (LSTM) network, and Bidirectional LSTM (Bi-LSTM), effectively capturing spatial and temporal dependencies within weather data crucial for accurate predictions. To optimize the DLM’s performance efficiently, a unique Kepler Optimization Algorithm (KOA) is introduced for hyperparameter tuning, drawing inspiration from Kepler’s laws of planetary motion. By leveraging KOA, the DLM attains optimal hyperparameter configurations, elevating power output prediction precision. Additionally, this study integrates Transductive Transfer Learning (TTL) with the deep learning models to enhance resource efficiency. By leveraging knowledge gained from previously learned tasks, TTL enables the DLM to improve its forecasting capabilities while minimizing resource utilization. Datasets encompassing environmental parameters and PV plant-generated power across diverse sites are employed for DLM training and testing. Three hybrid models, amalgamating KOA, CNN, LSTM, and Bi-LSTM techniques, are introduced and evaluated. Comparative assessment of these models across distinct PV sites yields insightful observations. Performance evaluation, focused on short-term PV power forecasting, underscores the hybrid DLM’s superiority over individual CNN and LSTM models. This hybrid approach achieves remarkable accuracy and resilience in predicting power output under varying weather conditions, showcasing its potential for efficient PV power plant management.

Multi-level Graph Memory Network Cluster Convolutional Recurrent Network for traffic forecasting

PSTFormer: A novel parallel spatial-temporal transformer for remaining useful life prediction of aeroengine

Spatial–temporal dependence and similarity aware traffic flow forecasting

This study uniquely integrates Long Short-Term Memory networks (LSTM) and Graph Convolutional Networks (GCN) with a multi-head attention mechanism to address dynamic carbon pricing optimization in green transportation supply chains, overcoming the limitations of traditional static models. As global climate change issues become increasingly severe, the design of carbon pricing mechanisms for green transportation supply chains has become a key factor in promoting sustainable development. We construct a hybrid deep learning model that simultaneously captures temporal dependencies in carbon emission data and spatial relationships in supply chain network structures. Traditional carbon pricing methods often rely on static models and simplified assumptions, making it difficult to adapt to complex and dynamic supply chain environments. Experimental results show that the proposed deep learning method improves carbon price prediction accuracy by 23.7% compared to traditional methods and achieves 18.5% improvement in supply chain cost optimization. Furthermore, the method achieved an average 21.6% carbon emission reduction and 15.5% cost reduction in three real green transportation supply chain cases, demonstrating its effectiveness in practical applications. The multi-objective optimization framework successfully balances the trade-off between economic and environmental benefits through organic integration of genetic algorithms and deep learning models. Ablation experiments validated the importance of each model component, and sensitivity analysis confirmed the rationality of parameter settings. This method provides strong technical support for formulating more precise and dynamic carbon pricing policies, offering significant theoretical value and practical significance for promoting sustainable development of green transportation supply chains.OPEN ACCESS Received: 17/09/2025 Accepted: 24/11/2025 Published: 20/03/2026

As the fundamental part of other Intelligent Transportation Systems (ITS) applications, short-term traffic volume prediction plays an important role in various intelligent transportation tasks, such as traffic management, traffic signal control and route planning. Although Neural-network-based traffic prediction methods can produce good results, most of the models can’t be explained in an intuitive way. In this paper, we not only proposed a model that increase the short-term prediction accuracy of the traffic volume, but also improved the interpretability of the model by analyzing the internal attention score learnt by the model. we propose a spatiotemporal attention mechanism-based multistep traffic volume prediction model (SAMM). Inside the model, an LSTM-based Encoder-Decoder network with a hybrid attention mechanism is introduced, which consists of spatial attention and temporal attention. In the first level, the local and global spatial attention mechanisms considering the micro traffic evolution and macro pattern similarity, respectively, are applied to capture and amplify the features from the highly correlated entrance stations. In the second level, a temporal attention mechanism is employed to amplify the features from the time steps captured as contributing more to the future exit volume. Considering the time-dependent characteristics and the continuity of the recent evolutionary traffic volume trend, the timestamp features and historical exit volume series of target stations are included as the external inputs. An experiment is conducted using data from the highway toll collection system of Guangdong Province, China. By extracting and analyzing the weights of the spatial and temporal attention layers, the contributions of the intermediate parameters are revealed and explained with knowledge acquired by historical statistics. The results show that the proposed model outperforms the state-of-the-art model by 29.51% in terms of MSE, 13.93% in terms of MAE, and 5.69% in terms of MAPE. The effectiveness of the Encoder-Decoder framework and the attention mechanism are also verified.

CPNet: Conditionally parameterized graph convolutional network for traffic forecasting

Accurate traffic speed forecasting is a prerequisite for anticipating future traffic status and increasing the resilience of intelligent transportation systems. However, most studies ignore the involvement of context information ubiquitously distributed over the urban environment to boost speed prediction. The diversity and complexity of context information also hinder incorporating it into traffic forecasting. Therefore, this study proposes a multimodal context-based graph convolutional neural network (MCGCN) model to fuse context data into traffic speed prediction, including spatial and temporal contexts. The proposed model comprises three modules, ie (a) hierarchical spatial embedding to learn spatial representations by organizing spatial contexts from different dimensions, (b) multivariate temporal modeling to learn temporal representations by capturing dependencies of multivariate temporal contexts and (c) attention-based multimodal fusion to integrate traffic speed with the spatial and temporal context representations for multi-step speed prediction. We conduct extensive experiments in Singapore. Compared to the baseline model (spatial-temporal graph convolutional network, STGCN), our results demonstrate the importance of multimodal contexts with the mean-absolute-error improvement of 0.29 km/h, 0.45 km/h and 0.89 km/h in 30-min, 60-min and 120-min speed prediction, respectively. We also explore how different contexts affect traffic speed forecasting, providing references for stakeholders to understand the relationship between context information and transportation systems.

A novel anomaly detection method for multivariate time series based on spatial-temporal graph learning

Traffic forecasting has emerged as an important task for developing intelligent transportation systems. Recent works focus on representing traffic as graph operation and using graph neural networks for spatial–temporal prediction. Most of the approaches assume a predefined graph structure based on node distances. However, spatial dependencies change over time in many scenarios of traffic flow. In this regard, this study takes an investigation capturing the spatial and temporal dependencies with no prior knowledge structure of traffic road networks. Specifically, we propose a multi-step prediction model named Dynamic Spatial Transformer WaveNet Network (DSTWN) to capture the dynamic conditions and directions of traffic flow in which a temporal convolution layer is adopted for the long time sequence and a spatial transformer layer is proposed to capture the dynamic spatial dependencies. Furthermore, we introduce a new traffic dataset, which is collected from the vehicle detection system in an urban area (UVDS). In particular, compared with existing benchmark traffic data, UVDS contains more complicated spatial information, which is similar to many real-world scenarios of traffic flow. Experiments on both benchmark traffic datasets indicate the promising results of DSTWN compared with state-of-the-art models in this research field.