Using the MapReduce Approach for the Spatio-Temporal Data Analytics in Road Traffic Crowdsensing Application

Abstract

Crowdsensing applications are becoming more popular with time. In this work, we present a crowdsensing application for capturing road traffic information to help citizens to get real-time traffic condition. Such real-time information can be beneficial for citizens to plan their journeys. However, crowdsensing in this specific case, generates spatio-temporal data collected from numerous users; storing and processing such data in real-time can be quite challenging. The MapReduce programming approach has been proposed for processing data in this context. The MapReduce jobs used to process and analyze the data captured from the crowdsensing application are presented as well as the design of the crowdsensing application. Implementation of the MapReduce jobs proposed shows that data can be effectively processed and analyzed to present near real-time information about the road traffic flow while at the same time discarding used data which is no longer required.