Short-Term Traffic Data Forecasting: A Deep Learning Approach

Agafonov, Anton

doi:10.3103/s1060992x21010021

Cited by 15 publications

(7 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Addressing this gap, the authors of refs. [ 16 , 17 ] introduced different timeframe durations, yet model performance diminishes in regions with varying demand densities.…”

Section: Literature Reviewmentioning

confidence: 99%

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction

Brahimi,

Zhang,

Zaidi

et al. 2024

Sensors

View full text Add to dashboard Cite

Car-sharing systems require accurate demand prediction to ensure efficient resource allocation and scheduling decisions. However, developing precise predictive models for vehicle demand remains a challenging problem due to the complex spatio-temporal relationships. This paper introduces USTIN, the Unified Spatio-Temporal Inference Prediction Network, a novel neural network architecture for demand prediction. The model consists of three key components: a temporal feature unit, a spatial feature unit, and a spatio-temporal feature unit. The temporal unit utilizes historical demand data and comprises four layers, each corresponding to a different time scale (hourly, daily, weekly, and monthly). Meanwhile, the spatial unit incorporates contextual points of interest data to capture geographic demand factors around parking stations. Additionally, the spatio-temporal unit incorporates weather data to model the meteorological impacts across locations and time. We conducted extensive experiments on real-world car-sharing data. The proposed USTIN model demonstrated its ability to effectively learn intricate temporal, spatial, and spatiotemporal relationships, and outperformed existing state-of-the-art approaches. Moreover, we employed negative binomial regression with uncertainty to identify the most influential factors affecting car usage.

show abstract

“…Addressing this gap, the authors of refs. [ 16 , 17 ] introduced different timeframe durations, yet model performance diminishes in regions with varying demand densities.…”

Section: Literature Reviewmentioning

confidence: 99%

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction

Brahimi,

Zhang,

Zaidi

et al. 2024

Sensors

View full text Add to dashboard Cite

show abstract

“…When the acquisition equipment has problems in any stage of data acquisition, processing and storage, the acquired traffic flow data set will inevitably have obvious defects. If these defects are allowed to flow into the algorithm model without necessary treatment, it will mislead the learning process of the algorithm model, seriously reduce the learning ability and greatly reduce the reliability of the model [19].Therefore, before the algorithm modeling, it is necessary to set the data preprocessing module [20]. The purpose of data preprocessing is to convert the original data input into high-quality input suitable for subsequent mining process, which usually includes integration, standardization, cleaning conversion and other technologies [21].…”

Section: Data Preprocessingmentioning

confidence: 99%

A Real Time Traffic Flow Model Based on Deep Learning

2022

KSII TIIS

View full text Add to dashboard Cite

Urban development has brought about the increasing saturation of urban traffic demand, and traffic congestion has become the primary problem in transportation. Roads are in a state of waiting in line or even congestion, which seriously affects people's enthusiasm and efficiency of travel. This paper mainly studies the discrete domain path planning method based on the flow data. Taking the traffic flow data based on the highway network structure as the research object, this paper uses the deep learning theory technology to complete the path weight determination process, optimizes the path planning algorithm, realizes the vehicle path planning application for the expressway, and carries on the deployment operation in the highway company. The path topology is constructed to transform the actual road information into abstract space that the machine can understand. An appropriate data structure is used for storage, and a path topology based on the modeling background of expressway is constructed to realize the mutual mapping between the two. Experiments show that the proposed method can further reduce the interpolation error, and the interpolation error in the case of random missing is smaller than that in the other two missing modes. In order to improve the real-time performance of vehicle path planning, the association features are selected, the path weights are calculated comprehensively, and the traditional path planning algorithm structure is optimized. It is of great significance for the sustainable development of cities.

show abstract

“…On the basis of deep learning, Son et al and Jin et al adopted the LSTM model to achieve spatiotemporal data prediction and conducted visual analysis [6,7]. Guo et al and Agafonov realized the prediction of microinternal leakage by analyzing the data of hydraulic cylinder based on the neural network model [8,9]. e above method has made some research results in data analysis and prediction.…”

Section: Related Workmentioning

confidence: 99%

“…e accuracy of the whole model is improved by trying to reduce the deviation of each decision tree. For the regression and classification problems, the decision trees adopted by the GBDT model are CART regression trees [9].…”

Section: Introduction To Gbdt Modelmentioning

confidence: 99%

Data Analysis and Prediction Modeling Based on Deep Learning in E-Commerce

Feng

2022

Scientific Programming

View full text Add to dashboard Cite

Due to the low efficiency of traditional data analysis methods for massive e-commerce data analysis, an e-commerce data analysis and prediction method based on the GBDT deep learning model was proposed. Purchase behavior is divided into another category, which transforms the problem of e-commerce data analysis and prediction into a binary classification problem. At the same time, we extract 107 features that can reflect the user behavior and construct the GBDT model. The characteristics include counting class, sorting class, time difference class, conversion rate class, and so on. It follows from the above that the analysis and prediction of e-commerce data are realized. In addition, the results show that when the learning rate of GBDT model parameters is 0.05, the number of basic learners is 200, the tree depth is 20, the threshold is 0.5, the model prediction effect is best, and the F1 value can reach 0.12. Compared with the traditional prediction model based on logistic regression and neural network, the proposed GBDT model is more suitable for e-commerce data analysis and prediction.

show abstract

Short-Term Traffic Data Forecasting: A Deep Learning Approach

Cited by 15 publications

References 25 publications

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction

A Unified Spatio-Temporal Inference Network for Car-Sharing Serial Prediction

A Real Time Traffic Flow Model Based on Deep Learning

Data Analysis and Prediction Modeling Based on Deep Learning in E-Commerce

Contact Info

Product

Resources

About