Missing traffic data imputation and pattern discovery with a Bayesian augmented tensor factorization model

Chen, Xinyu; He, Zhaocheng; Chen, Yixian; Lu, Yuhuan; Wang, Jiawei

doi:10.1016/j.trc.2019.03.003

Cited by 113 publications

(36 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The matrix-decomposition-based data reconstruction method (e.g., low-rank matrix/tensor completion) [55] is one of the widely advocated algorithms due to the direct expression of the relationship between time series and spatial locations. Tensor completion methods can be used in different fields to solve the missing data problem and are generally used for image completion and traffic data imputation [59][60][61][62][63][64]. However, investigations on their application to the data imputation of three-axial coupled structural responses of buildings under seismic excitation for building safety assessment are rarely reported.…”

Section: Methods For Fast Missing Data Reconstructionmentioning

confidence: 99%

Study of Building Safety Monitoring by Using Cost-Effective MEMS Accelerometers for Rapid After-Earthquake Assessment with Missing Data

Lin

Wang

et al. 2021

Sensors

View full text Add to dashboard Cite

Suffering from structural deterioration and natural disasters, the resilience of civil structures in the face of extreme loadings inevitably drops, which may lead to catastrophic structural failure and presents great threats to public safety. Earthquake-induced extreme loading is one of the major reasons behind the structural failure of buildings. However, many buildings in earthquake-prone areas of China lack safety monitoring, and prevalent structural health monitoring systems are generally very expensive and complicated for extensive applications. To facilitate cost-effective building-safety monitoring, this study investigates a method using cost-effective MEMS accelerometers for buildings’ rapid after-earthquake assessment. First, a parameter analysis of a cost-effective MEMS sensor is conducted to confirm its suitability for building-safety monitoring. Second, different from the existing investigations that tend to use a simplified building model or small-scaled frame structure excited by strong motions in laboratories, this study selects an in-service public building located in a typical earthquake-prone area after an analysis of earthquake risk in China. The building is instrumented with the selected cost-effective MEMS accelerometers, characterized by a low noise level and the capability to capture low-frequency small-amplitude dynamic responses. Furthermore, a rapid after-earthquake assessment scheme is proposed, which systematically includes fast missing data reconstruction, displacement response estimation based on an acceleration response integral, and safety assessment based on the maximum displacement and maximum inter-story drift ratio. Finally, the proposed method is successfully applied to a building-safety assessment by using earthquake-induced building responses suffering from missing data. This study is conducive to the extensive engineering application of MEMS-based cost-effective building monitoring and rapid after-earthquake assessment.

show abstract

Section: Methods For Fast Missing Data Reconstructionmentioning

confidence: 99%

Study of Building Safety Monitoring by Using Cost-Effective MEMS Accelerometers for Rapid After-Earthquake Assessment with Missing Data

Lin

Wang

et al. 2021

Sensors

View full text Add to dashboard Cite

show abstract

“…In the application of a data-driven traffic flow prediction model, it is inevitable that we will encounter the missing data problem, which usually leads to incorrect predictions and responses (X. Chen et al, 2019). To further explore the robustness of the proposed STGGAT model to perturbations, we insert random noise and stochastic missing data into the LPR dataset and validate the fault tolerance of STGGAT.…”

Section: Fault Tolerance Analysismentioning

confidence: 99%

Spatiotemporal gated graph attention network for urban traffic flow prediction based on license plate recognition data

Tang

Zeng

2021

Computer aided Civil Eng

View full text Add to dashboard Cite

The accurate forecasting of traffic states is an essential application of intelligent transportation system. Due to the periodic signal control at intersections, the traffic flow in an urban road network is often disturbed and expresses intermittent features. This study proposes a forecasting framework named the spatiotemporal gated graph attention network (STGGAT) model to achieve accurate predictions for network-scale traffic flows on urban roads. Based on license plate recognition (LPR) records, the average travel times and volume transition relationships are estimated to construct weighted directed graphs. The proposed STGGAT model integrates a gated recurrent unit layer, a graph attention network layer with edge features, a gated mechanism based on the bidirectional long short-term memory and a residual structure to extract the spatiotemporal dependencies of the approach-and lane-level traffic volumes. Validated on the LPR system in Changsha, China, STGGAT demonstrates superior accuracy and stability to those of the baselines and reveals its inductive learning and fault tolerance capabilities.How to cite this article: Tang J, Zeng J. Spatiotemporal gated graph attention network for urban traffic flow prediction based on license plate recognition data.

show abstract

“…Recently, tensor decomposition has been extensively used to recover missing data in different fields. For example, missing traffic data was recovered in [25] using Bayesian augmented tensor factorization model. They exploited Bayesian framework for automatically learning parameters of this model using variational Bayes.…”

Section: A Related Workmentioning

confidence: 99%

Classification Analysis of Tensor-Based Recovered Missing EEG Data

Akmal

Zubair²,

Alquhayz

2021

IEEE Access

View full text Add to dashboard Cite

This paper discusses impact of recovering missing Electroencephalography (EEG) data on classification accuracy of hand movements using tensor-based methods. Improvement in classification accuracy is important for efficient performance of prosthesis. Classification accuracy relies on quality of the observed data. In practice, observed data is usually incomplete because of disconnection of electrodes and other artefacts, which negatively affect the classification accuracy. In this paper, we employ tensor-based imputation methods (canonical/polyadiac decomposition (CPD), weighted optimization version of CPD (CP-WOPT and Nonnegative Matrix Factorization (NMF)) to recover missing data in EEG signals and apply various classifiers to classify hand movements on recovered data. In particular, structured missing data was considered because that is how data gets missed in real-life data acquisition. Percentage of missing data was changed from 10% to 50%. Classifiers are applied on complete, missing and recovered data explicitly to test the performance of our framework. Results show that mean classification accuracy on complete data, missing data and recovered data was 71%, 53% and 64% respectively which shows applicability of our framework.

show abstract

Missing traffic data imputation and pattern discovery with a Bayesian augmented tensor factorization model

Cited by 113 publications

References 13 publications

Study of Building Safety Monitoring by Using Cost-Effective MEMS Accelerometers for Rapid After-Earthquake Assessment with Missing Data

Study of Building Safety Monitoring by Using Cost-Effective MEMS Accelerometers for Rapid After-Earthquake Assessment with Missing Data

Spatiotemporal gated graph attention network for urban traffic flow prediction based on license plate recognition data

Classification Analysis of Tensor-Based Recovered Missing EEG Data

Contact Info

Product

Resources

About