Long-term traffic flow forecasting using a hybrid CNN-BiLSTM model

Méndez, Manuel; Merayo, Mercedes G.; Núñez, Manuel

doi:10.1016/j.engappai.2023.106041

Cited by 80 publications

(9 citation statements)

References 37 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The BiLSTM model effectively captures longterm dependencies in sequential data, while the CNN model excels at learning local patterns in the input. We then utilized the attention mechanism to highlight important features in the input data, improving the model's focus on relevant opcode sequences that contribute to more comprehensive vulnerability detection [34,39]. The subsequent equations detail the mathematical foundations of the BiLSTM, CNN, and attention mechanisms employed in the proposed model.…”

Section: Data Preprocessingmentioning

confidence: 99%

VdaBSC: A Novel Vulnerability Detection Approach for Blockchain Smart Contract by Dynamic Analysis

Sosu,

Chen,

Boahen

et al. 2023

IET Software

View full text Add to dashboard Cite

Smart contracts have gained immense popularity in recent years as self-executing programs that operate on a blockchain. However, they are not immune to security flaws, which can result in significant financial losses. These flaws can be detected using dynamic analysis methods that extract various aspects from smart contract bytecode. Methods currently used for identifying vulnerabilities in smart contracts mostly rely on static analysis methods that search for predefined vulnerability patterns. However, these patterns often fail to capture complex vulnerabilities, leading to a high rate of false negatives. To overcome this limitation, researchers have explored machine learning-based methods. However, the accurate interpretation of complex logic and structural information in smart contract code remains a challenge. In this study, we present a technique that combines real-time runtime batch normalization and data augmentation for data preprocessing, along with n-grams and one-hot encoding for feature extraction of opcode sequence information from the bytecode. We then combined bidirectional long short-term memory (BiLSTM), convolutional neural network, and the attention mechanism for vulnerability detection and classification. Additionally, our model includes a gated recurrent units memory module that enhances efficiency using historical execution data from the contract. Our results demonstrate that our proposed model effectively identifies smart contract vulnerabilities.

show abstract

Section: Data Preprocessingmentioning

confidence: 99%

VdaBSC: A Novel Vulnerability Detection Approach for Blockchain Smart Contract by Dynamic Analysis

Sosu,

Chen,

Boahen

et al. 2023

IET Software

View full text Add to dashboard Cite

show abstract

“…Convolutional neural networks (CNNs) have also been applied for modeling spatialtemporal traffic data to capture the temporal dependencies or spatial correlations in traffic networks [18][19][20][21]. However, traditional CNNs are often limited to modeling Euclidean data, and graph convolutional networks (GCNs) have gained popularity since 2018, as traffic data usually have non-Euclidean spatial structures and can be represented in graph form.…”

Section: Introductionmentioning

confidence: 99%

Adaptive Graph Attention and Long Short-Term Memory-Based Networks for Traffic Prediction

Zhu,

Boada,

Boada

2024

Mathematics

View full text Add to dashboard Cite

While the increased availability of traffic data is allowing us to better understand urban mobility, research on data-driven and predictive modeling is also providing new methods for improving traffic management and reducing congestion. In this paper, we present a hybrid predictive modeling architecture, namely GAT-LSTM, by incorporating graph attention (GAT) and long short-term memory (LSTM) networks for handling traffic prediction tasks. In this architecture, GAT networks capture the spatial dependencies of the traffic network, LSTM networks capture the temporal correlations, and the Dayfeature component incorporates time and external information (such as day of the week, extreme weather conditions, holidays, etc.). A key attention block is designed to integrate GAT, LSTM, and the Dayfeature components as well as learn and assign weights to these different components within the architecture. This method of integration is proven effective at improving prediction accuracy, as shown by the experimental results obtained with the PeMS08 open dataset, and the proposed model demonstrates state-of-the-art performance in these experiments. Furthermore, the hybrid model demonstrates adaptability to dynamic traffic conditions, different prediction horizons, and various traffic networks.

show abstract

“…This work chronicles the methodical development of an intelligent short-and long-term urban traffic prediction system [14], [15], encompassing various mobility data types, traffic modeling techniques, and critical traffic indicators such as speed, flow, and accident risk [16], [17]. Special emphasis is placed on time series analysis and the pivotal role of data preprocessing, encompassing normalization, transformation, outlier handling, and feature engineering, crucial in enhancing the predictive accuracy of deep learning models [18].…”

Section: Introductionmentioning

confidence: 99%

Elevating smart city mobility using RAE-LSTM fusion for next-gen traffic prediction

Rafalia,

Moumen,

Raji

et al. 2024

IJEECS

View full text Add to dashboard Cite

The burgeoning demand for efficient urban traffic management necessitates accurate prediction of traffic congestion, spotlighting the essence of time series data analysis. This paper delves into the utilization of sophisticated deep learning methodologies, particularly long short-term memory (LSTM) networks, convolutional neural networks (CNN), and their amalgamations like Conv-LSTM and bidirectional-LSTM (Bi-LSTM), to elevate the precision of traffic pattern forecasting. These techniques showcase promise in encapsulating the intricate dynamics of traffic flow, yet their efficacy hinges upon the quality of input data, emphasizing the pivotal role of data preprocessing. This study meticulously investigates diverse preprocessing techniques encompassing normalization, transformation, outlier detection, and feature engineering. Its discerning implementation significantly heightens the performance of deep learning models. By synthesizing advanced deep learning architectures with varied preprocessing methodologies, this research presents invaluable insights fostering enhanced accuracy and reliability in traffic prediction. The innovative RD-LSTM approach introduced herein harnesses the hybridization of a reverse AutoEncoder and LSTM models, marking a novel contribution to the field. The implementation of these progressive strategies within urban traffic management portends substantial enhancements in efficiency and congestion mitigation. Ultimately, these advancements pave the way for a superior urban experience, enriching the quality of life within cities through optimized traffic management systems.

show abstract

Long-term traffic flow forecasting using a hybrid CNN-BiLSTM model

Cited by 80 publications

References 37 publications

VdaBSC: A Novel Vulnerability Detection Approach for Blockchain Smart Contract by Dynamic Analysis

VdaBSC: A Novel Vulnerability Detection Approach for Blockchain Smart Contract by Dynamic Analysis

Adaptive Graph Attention and Long Short-Term Memory-Based Networks for Traffic Prediction

Elevating smart city mobility using RAE-LSTM fusion for next-gen traffic prediction

Contact Info

Product

Resources

About