Traffic Prediction Using Gated Recurrent Unit Neural Networks

Abstract

Abstract— In an intelligent transportation system, traffic prediction is vital. Accurate traffic forecasting can help with route selection, vehicle dispatching, and traffic congestion reduction. Due to the complex and dynamic spatio-temporal relationships between different parts in the road network, this problem is difficult to solve. Recently, a large amount of research work has been committed to this area, particularly the machine learning method, which has substantially improved traffic forecast abilities. Despite the fact that the infrastructure is outdated and can only support a small population, there is an influx of residents looking for work and opportunity. Fuel combustion is enhanced as a result of traffic congestion. In this project, i will be able to be exploring the dataset of 4 junctions and built a model to predict traffic on an equivalent . This could potentially help in solving the traffic jam problem by providing a far better understanding of traffic patterns which will further help in building an infrastructure to eliminate the matter .