Long Short-Term Memory Networks for Traffic Flow Forecasting: Exploring Input Variables, Time Frames and Multi-Step Approaches

A major challenge of today’s society is to make large urban centres more sustainable. Improving the energy efficiency of the various infrastructures that make up cities is one aspect being considered when improving their sustainability, with Wastewater Treatment Plants (WWTPs) being one of them. Consequently, this study aims to conceive, tune, and evaluate a set of candidate deep learning models with the goal being to forecast the energy consumption of a WWTP, following a recursive multi-step approach. Three distinct types of models were experimented, in particular, Long Short-Term Memory networks (LSTMs), Gated Recurrent Units (GRUs), and uni-dimensional Convolutional Neural Networks (CNNs). Uni- and multi-variate settings were evaluated, as well as different methods for handling outliers. Promising forecasting results were obtained by CNN-based models, being this difference statistically significant when compared to LSTMs and GRUs, with the best model presenting an approximate overall error of 630 kWh when on a multi-variate setting. Finally, to overcome the problem of data scarcity in WWTPs, transfer learning processes were implemented, with promising results being achieved when using a pre-trained uni-variate CNN model, with the overall error reducing to 325 kWh.

show abstract

“…In this example, the labels would have the shape (1601, 1). A similar algorithm can be seen in the work of Fernandes et al [15].…”

Section: Supervised Problemmentioning

confidence: 83%

Forecasting Energy Consumption of Wastewater Treatment Plants with a Transfer Learning Approach for Sustainable Cities

et al. 2021

Self Cite

View full text Add to dashboard Cite

show abstract

“…By using the minimum maximum normalization method, the values of the raw data are mapped to the range [0,1]. Traffic flow prediction (TFP) includes short and long-term ones, and the former is the key to alleviating traffic congestion [49][50]. The short-term TFP model based on AM and CNN-LSTM is shown in Figure 4.…”

Section: B Design Of Stop Point Recognition and Construction Of Dtprp...mentioning

confidence: 99%

Construction of Dynamic Traffic Pattern Recognition and Prediction Model Based on Deep Learning in the Background of Intelligent Cities

Cui,

Zhao

2024

IEEE Access

View full text Add to dashboard Cite

To reduce traffic congestion, it is particularly important to use advanced technology to predict urban traffic flow. Therefore, a dynamic traffic pattern prediction model is proposed, which includes convolutional neural network, long and short term memory network and attention mechanism. The validity of the prediction model is verified by the loss function and the average absolute percentage error. In addition, the study also constructs a model for user travel pattern and parking point recognition based on deep learning and mobile signaling data. The performance of the recognition model is verified by the accuracy and other indicators. The research outcomes demonstrated that the max average absolute percentage error of the dynamic traffic mode prediction model was 7.8%, and the mini value was 2.9%. The average accuracy of the user travel pattern recognition model was 83.34%, and that of the parking point recognition model was 88.56%.The dynamic traffic model recognition and prediction model designed by the research institute has better results, and has practical guiding significance in smart city traffic management.

show abstract

“…However, these applications are not able to logically explain their autonomous decisions and actions to human users. Although explanations may not be essential for specific applications, for many critical applications, such as agriculture and environmental projects [1][2][3][4], traffic-flow management and object detection [5,6], and ailment cues [7], explanations are essential for users to understand, trust, and effectively manage these new artificially intelligent partners [8].…”

Section: Introductionmentioning

confidence: 99%

Neural Network Explainable AI Based on Paraconsistent Analysis: An Extension

Marcondes¹,

Durães²,

Santos³

et al. 2021

Electronics

Self Cite

View full text Add to dashboard Cite

This paper explores the use of paraconsistent analysis for assessing neural networks from an explainable AI perspective. This is an early exploration paper aiming to understand whether paraconsistent analysis can be applied for understanding neural networks and whether it is worth further develop the subject in future research. The answers to these two questions are affirmative. Paraconsistent analysis provides insightful prediction visualisation through a mature formal framework that provides proper support for reasoning. The significant potential envisioned is the that paraconsistent analysis will be used for guiding neural network development projects, despite the performance issues. This paper provides two explorations. The first was a baseline experiment based on MNIST for establishing the link between paraconsistency and neural networks. The second experiment aimed to detect violence in audio files to verify whether the paraconsistent framework scales to industry level problems. The conclusion shown by this early assessment is that further research on this subject is worthful, and may eventually result in a significant contribution to the field.

show abstract

Long Short-Term Memory Networks for Traffic Flow Forecasting: Exploring Input Variables, Time Frames and Multi-Step Approaches

Cited by 12 publications

References 34 publications

Forecasting Energy Consumption of Wastewater Treatment Plants with a Transfer Learning Approach for Sustainable Cities

Forecasting Energy Consumption of Wastewater Treatment Plants with a Transfer Learning Approach for Sustainable Cities

Construction of Dynamic Traffic Pattern Recognition and Prediction Model Based on Deep Learning in the Background of Intelligent Cities

Neural Network Explainable AI Based on Paraconsistent Analysis: An Extension

Contact Info

Product

Resources

About