Performance improvement for metro passenger flow forecast using spatio-temporal deep neural network

Mulerikkal, Jaison; Thandassery, Sajanraj; Rejathalal, Vinith; Kunnamkody, Deepa Merlin Dixon

doi:10.1007/s00521-021-06522-5

Cited by 13 publications

(6 citation statements)

References 21 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This is understandable by the fact that typical time series prediction techniques such as AR, MA, WA), ARIMA, etc., don't leverage multivariate data inter-dependencies and are usually "same-to-same" feature prediction. Among common outlier filtering techniques for univariate data, we find z-score filtering [12], one-class Support Vector Machine (SVM) [13]- [15] and Local Outlier Factor (LOF) [16], [17].…”

Section: B Outlier Filtering For Data Pre-processingmentioning

confidence: 99%

“…Approaches in [13]- [15] use one-class SVM for outlier data detection, then elimination because this approach is simple, efficient and can guarantee the authenticity of data to a certain extent. One-class SVM can greatly improve the precision of anomaly detection in the case of small samples, unbalanced sample classification, and supposes no assumptions about data distribution.…”

Section: B Outlier Filtering For Data Pre-processingmentioning

confidence: 99%

“…One-class SVM can greatly improve the precision of anomaly detection in the case of small samples, unbalanced sample classification, and supposes no assumptions about data distribution. For example, the authors in [13] do so to reduce irregular patterns in metro passenger flow forecasting from an LSTM network. Next, the authors in [14] introduce a novel two-step LSTM configuration for removing deterministic components associated with dominant gear signals in the first step (with LSTM regression) and removing the 'residual deterministic' components associated with varying gear signals in the second step (with one-class SVM).…”

Section: B Outlier Filtering For Data Pre-processingmentioning

confidence: 99%

See 2 more Smart Citations

Multivariate outlier filtering for A-NFVLearn: an advanced deep VNF resource usage forecasting technique

2023

View full text Add to dashboard Cite

Section: B Outlier Filtering For Data Pre-processingmentioning

confidence: 99%

Section: B Outlier Filtering For Data Pre-processingmentioning

confidence: 99%

Section: B Outlier Filtering For Data Pre-processingmentioning

confidence: 99%

See 1 more Smart Citation

Multivariate outlier filtering for A-NFVLearn: an advanced deep VNF resource usage forecasting technique

2023

View full text Add to dashboard Cite

“…The goal of deep network training is to maximize (or minimize) E (x) through effective training mode. The accuracy of training mainly includes the design of E (x) and initialization of internal parameters of the model [13][14]. The function of gradient descent is to find the minimum value, control the variance, update the model parameters, and finally make the model converge:…”

Section: Deep Network Improvementmentioning

confidence: 99%

Improvement of Diesel Engine Cooling System Based on Deep Learning Algorithm

2020

KME

View full text Add to dashboard Cite

The core idea of deep learning is to process input vectors and take their outputs as training samples, and then use classifiers to predict and modify these input features. When improving the existing diesel engine cooling system, we can use deep learning to solve the problem. Therefore, the purpose of this paper is to improve the design of diesel engine cooling system and improve the working efficiency. In this paper, the experimental method and comparison method are mainly used to build a diesel engine test bench. By setting the inlet water temperature of the diesel engine, the variation rules and influencing factors of the cooling water temperature and flow rate under three temperatures are analyzed. The experimental results show that when the inlet water temperature of the diesel engine is 90 ℃, the cooling water flow always takes the first place, which indicates that the diesel engine has good performance under this condition.

show abstract

“…Deep learning models have been shown to outperform traditional methods as a branch of machine learning in capturing complex correlations from big data and researchers have gradually attempted to apply models in these fields to passenger flow prediction as a branch of machine learning [12]. Given the good performance of time series forecasting, recurrent neural network (RNN) and their variants, such as long short-term memory (LSTM) [13][14][15] and gated recurrent unit (GRU), are widely used in mainstream research on traffic forecasting [16,17]. However, RNN-based models only exploit the temporal features in travel behaviour, while ignoring the potential spatial correlations in the transportation network.…”

Section: Introductionmentioning

confidence: 99%

A deep neural network model with GCN and 3D convolutional network for short‐term metro passenger flow forecasting

Zhang

Wang

Chen

et al. 2023

IET Intelligent Trans Sys

View full text Add to dashboard Cite

Rail transit has many advantages, such as large passenger capacity, convenience, safety, and environmental protection, making it the preferred travel mode for most passengers. Deep learning has become an effective method for short‐term rail passenger flow prediction. A deep learning model which combines a graph convolutional network and a three‐dimensional Convolutional Neural Network improved by a residual module and an attention mechanism (ARConv‐CGN) is proposed. First, historical passenger inflow and outflow data are aggregated into three patterns: recent pattern, daily pattern, and weekly pattern separately. The GCN is applied to capture the spatiotemporal and topological information of passenger flows in each pattern. Second, a three‐dimensional convolutional neural network is used to deeply integrate three patterns of passenger flow information. Additionally, a residual module is used to increase the number of neural network layers and prevent the gradient of the deep neural network from disappearing. Finally, an attention mechanism is also introduced to adjust the importance of passenger flow in different pattern, so that improving the performance of the model. Training with passenger flow data from automatic fare collection on weekdays in Beijing and Xiamen provides a good demonstration of the superiority of ARConv‐CGN in metro passenger flow forecasting.

show abstract

Performance improvement for metro passenger flow forecast using spatio-temporal deep neural network

Cited by 13 publications

References 21 publications

Multivariate outlier filtering for A-NFVLearn: an advanced deep VNF resource usage forecasting technique

Multivariate outlier filtering for A-NFVLearn: an advanced deep VNF resource usage forecasting technique

Improvement of Diesel Engine Cooling System Based on Deep Learning Algorithm

A deep neural network model with GCN and 3D convolutional network for short‐term metro passenger flow forecasting

Contact Info

Product

Resources

About