RTIC-C: A Big Data System for Massive Traffic Information Mining

Abstract

Traffic information system may produce massive and complex traffic data with the process of collecting real-time original GPS (Global Positioning System) data, matching positions to a map and generating traffic flow information, which brings great markets for even-worse traffic condition in China. Howerver, several issues would occur when we reuse these massive traffic data for history data mining applying on-hand database management tools or traditional data processing apporaches, such as massive storage, high performance processing, open interface. "Big Data" system usually includes data sets with sizes beyond the ability of commonly-used software tools to capture, manage, and process the data within a tolerable elapsed time. With this difficulty and the advantage of "Big Data", we schemed RTIC-C system to handle sensemaking over large quantities of traffic data based on cloud computing technique. RTIC-C designs a distributed data management service to support large scale of data storage, a parallel distributed computing framework for diverse kinds of mining applications based on Map-Reduce mechanism, a restful Web services interface to support third-party mining applications. Experiments on a massive traffic data sets showed that RTIC-C achieves considerable performance comparing with traditional traffic data mining applications.