Prediction of Short Term Traffic Speeds Using Deep Learning Models
No Thumbnail Available
Date
2024
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
Institute of Electrical and Electronics Engineers Inc.
Open Access Color
OpenAIRE Downloads
OpenAIRE Views
Abstract
In this study, real-time traffic forecasting was conducted using traffic speed data obtained from Bluetooth sensors located in the city of Vigo, Spain. For this purpose, average speed data recorded at fifteen-minute intervals from five different sensors in the city since 2014 were used. Using past average speed values in a time series format as inputs, various deep learning methods were applied to predict traffic speed data for the next fifteen minutes. The Long Short-Term Memory (LSTM) method achieved the highest performance in traffic forecasting. Incorporating additional factors such as time, weather conditions, and environmental factors into the models, along with time-series traffic volumes, could further enhance the performance of near-future traffic forecasting. © 2024 IEEE.
Description
IEEE SMC; IEEE Turkiye Section
Keywords
Deep Learning, Long Short Term Memory, Traffic Prediction
Turkish CoHE Thesis Center URL
Fields of Science
Citation
WoS Q
N/A
Scopus Q
N/A
Source
2024 Innovations in Intelligent Systems and Applications Conference, ASYU 2024 -- 2024 Innovations in Intelligent Systems and Applications Conference, ASYU 2024 -- 16 October 2024 through 18 October 2024 -- Ankara -- 204562