Implementation of an application mobile for the traffic prediction

Abstract

In this theis, we proposed a new deep learning model to predict travel time based on GPS data collected in the city of Adrar. Our approach is based on two essential parts: The first part focuses on the data collection. In this phase, an android application, called GPS Adrar, is developed for the purpose of collecting GPS coordinates from Adrar’s citizens. The second part concerns data processing. In this phase, all the data collected in the first part must be analyzed using a depp learning model. Next, the new model is integrated with an android application, called Wassalni that offers the user a better prediction of road traffic among a set of available destinations. The experimental results show that the proposed model offers high accuracy than Google Maps in most of the routes.