Intelligent Request Grabber: Increases the Vehicle Traffic Prediction Rate Using Social and Taxi Requests Based on LSTM

Abstract

The integration of transportation system with Internet of Things (IoT) aims to effectively predict the vehicle traffic on a particular location. To optimize the transportation resources and learn about the public transportation system in the near future based on the transportation demand. We analyze the effect of machine intelligence on the transportation system; propose a new framework to improve the prediction accuracy on traffic. Mostly, transportation scheduling system based on historical data on usual routes, present system never involve realistic traffic situation or prolong changes on the system. This research work provides an efficient solution to avoid traffic and optimize the wastage of transportation resources. Social transportation data is pioneering attempt to acquire the very useful location based information that helps to understand the density of the human on geographical regions with mobility. In order to predict the real time traffic we collects the taxi request and convert the best sequence learning model, to predict continues data and analyze the dynamic traffic pattern in efficient manner. Experimental results achieve higher prediction accuracy of 92.4%.