Multi-attention graph neural networks for city-wide bus travel time estimation using limited data

J Ma, J Chan, S Rajasegarar, C Leckie - Expert Systems with Applications, 2022 - Elsevier
… We propose a bus network construction method from a novel … representation method
represents the road segments of bus … to predict city-wide bus travel times that considering spatial, …

Bus-arrival time prediction using bus network data model and time periods

M Čelan, M Lep - Future Generation Computer Systems, 2020 - Elsevier
… the historical data, statistical models, Kalman filtering models and machine learning models.
… This paper proposes an approach for predicting bus arrivals to the stop based on historical …

Does the Inclusion of Spatio-Temporal Features Improve Bus Travel Time Predictions? A Deep Learning-Based Modelling Approach

G Lee, S Choo, S Choi, H Lee - Sustainability, 2022 - mdpi.com
… Additionally, since a linear filter is usually used in the Kalman filter (KF) model, it was
difficult to … Therefore, in this study, the spatial features of bus stops are captured through bus

Bus travel times prediction based on real-time traffic data forecast using artificial neural networks

GH Larsen, LR Yoshioka… - … Conference on Electrical …, 2020 - ieeexplore.ieee.org
… We showed that our method could provide an accurate bus travel time prediction from web
data … Moreover, [31] applied a Kalman Filter to optimize the arrival time prediction of the …

Comparing two hybrid neural network models to predict real-world bus travel time

T Nimpanomprasert, L Xie, N Kliewer - Transportation Research Procedia, 2022 - Elsevier
… • We develop a new hybrid model taking a Kalman filter in an … The free space of the new
population is fulfilled with a new … A suitable approach for evaluating bus arrival time prediction

[HTML][HTML] Particle swarm optimization and RBF neural networks for public transport arrival time prediction using GTFS data

E Chondrodima, H Georgiou, N Pelekis… - International Journal of …, 2022 - Elsevier
… This work introduces a novel data-driven approach for predictingbus arrival time by using
automatic vehicle location data and by … for exploring the space of solutions by encoding each …

Dynamic origin–destination matrix prediction with line graph neural networks and kalman filter

X Xiong, K Ozbay, L Jin, C Feng - Transportation Research …, 2020 - journals.sagepub.com
Kalman filter to predict spatial-temporal OD flows is proposed. Since different topology matrices
are incorporated into graph neural networks and Kalman filter… of the proposed approach. …

Bus travel time prediction based on ensemble learning methods

G Zhong, T Yin, L Li, J Zhang… - IEEE Intelligent …, 2020 - ieeexplore.ieee.org
… set are applied in the prediction approaches according to the … analyze the prediction results
in time and space dimensions to … Qin, “Hybrid dual Kalman filtering model for short-term traffic …

Spatio-temporal modelling and prediction of bus travel time using a higher-order traffic flow model

D Bharathi, L Vanajakshi, SC Subramanian - Physica A: Statistical …, 2022 - Elsevier
… the state-state-space form and integrated with a filtering technique using appropriate inputs,
… LWR model-based prediction method. The developed real-time prediction methodology is a …

Intelligent Passenger Frequency Prediction for Transportation Sustainability using Kalman Filter Algorithm and Convolutional Neural Network

OD Jimoh, LA Ajao, OO Adeleke, SS Kolo… - 2022 - preprints.org
… considered only local spatial correlations due to the LSTM network method adopted. Cats …
prediction analysis for the dynamic time-table scheduling to manage the inter-campuses bus