کنفرانس بین المللی فناوری اطلاعات و دانش

English

صفحه اصلی / دوازدهمین کنفرانس بین المللی فناوری اطلاعات و دانش

Short-Term Traffic Flow Prediction Based on a Recurrent Deep Neural Networks: Study in Tehran

نویسندگان :

Monireh عبدوس¹ Taha Vajed Samei²

1- شهید بهشتی 2- shahid beheshti university

کلمات کلیدی :

Intelligent transportation systems, Urban traffic prediction, Long short term memory, Deep learning.

چکیده :

With growing of population, the issue of optimal mobility between two points of the city has become one of the most important problems. There are various tools to suggest the optimal route, but due to the momentary changes in traffic in cities, especially large cities, providing the optimal route without predicting the traffic load will not be accurate. In this regard, it can be noted that one of the most widely used up-to-date methods is the use of deep neural networks to predict the future. In this paper, while examining some of the most widely used deep neural networks to predict traffic sequence, a method is presented based on one of recurrent neural networks. The method has been evaluated on real traffic data on a part of Tehran. The results show that the proposed method outperforms the other similar neural networks.