<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>The urban intelligent transportation travel service system can effectively improve the quality of transportation travel, but the current system has certain deficiencies. This paper from two aspects to improve and analyze. On the one hand, the point-level artificial features are modeled through relevant mathematical models, and the citizens’ travel feature extraction model is designed based on a stack self-encoder after smoothing the artificial features. On the other hand, a traffic flow prediction model was designed through the temporal and spatial features of urban traffic, respectively, and combined with the spatio-temporal correlation feature design to predict the urban traffic flow. In addition, the effects of travel mode extraction and traffic flow prediction are analyzed to explore the optimization effect of the travel system. The results show that when the time value is taken as the 90s, the corresponding coding sparsity and weight are obtained as 50 and 0.05, respectively, and the average accuracy and average recall of the traffic modes reach 0.9. When the number of neurons is set to 30, the gap of the traffic flow in the peak is about 80, the gap of the traffic flow in the rest of the time period is about 10, and the relative error value is within 0.1. This study is able to optimize the intelligent traffic travel service system.</p>
</abstract>

The urban intelligent transportation travel service system can effectively improve the quality of transportation travel, but the current system has certain deficiencies. This paper from two aspects to improve and analyze. On the one hand, the point-level artificial features are modeled through relevant mathematical models, and the citizens’ travel feature extraction model is designed based on a stack self-encoder after smoothing the artificial features. On the other hand, a traffic flow prediction model was designed through the temporal and spatial features of urban traffic, respectively, and combined with the spatio-temporal correlation feature design to predict the urban traffic flow. In addition, the effects of travel mode extraction and traffic flow prediction are analyzed to explore the optimization effect of the travel system. The results show that when the time value is taken as the 90s, the corresponding coding sparsity and weight are obtained as 50 and 0.05, respectively, and the average accuracy and average recall of the traffic modes reach 0.9. When the number of neurons is set to 30, the gap of the traffic flow in the peak is about 80, the gap of the traffic flow in the rest of the time period is about 10, and the relative error value is within 0.1. This study is able to optimize the intelligent traffic travel service system.

Optimization of urban intelligent transportation travel service system based on feature extraction and traffic prediction

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

The urban intelligent transportation travel service system can effectively improve the quality of transportation travel, but the current system has certain...

Optimization of urban intelligent transportation travel service system based on feature extraction and traffic prediction

Publié en ligne: 29 nov. 2023

Pages: -

Reçu: 02 janv. 2023

Accepté: 09 juin 2023

DOI: https://doi.org/10.2478/amns.2023.2.01280

Mots clés
Stack self-encoder, Feature extraction, spatio-temporal correlation features, Traffic flow prediction, Traffic travel service

© 2023 Chunhong He, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Optimization of urban intelligent transportation travel service system based on feature extraction and traffic prediction

Publié en ligne: 29 nov. 2023

Pages: -

Reçu: 02 janv. 2023

Accepté: 09 juin 2023

DOI: https://doi.org/10.2478/amns.2023.2.01280

Mots clésStack self-encoder, Feature extraction, spatio-temporal correlation features, Traffic flow prediction, Traffic travel service

© 2023 Chunhong He, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Mots clés
Stack self-encoder, Feature extraction, spatio-temporal correlation features, Traffic flow prediction, Traffic travel service