Title: Methods for Gathering Road Big Data for Predicting the Driving Environment

Year of Publication: Oct - 2015
Page Numbers: 71-77
Authors: Jung-Ah Ha , Youn-Mi Jang, Kyu-Soo Chong and Je-Yoon Woo
Conference Name: The International Conference on Computer, Electronics, and Biomedical Engineering (ICCEBE2015)
- United Arab Emirates


This study examined methods designed to gather and store the diverse information on roads for predicting the driving environment. To formulate a plan for gathering road information with the aim of predicting the driving environment, the road information was classified into traffic information, traffic control information, road incident information, road status information, climate information, road management information, Prove information, and vehicle detection information. First, the types, contents, and sizes of road big data gathered internally and externally were defined, and data were gathered in real time and in non-real time. In the non-real-time data gathering, many data were brought from those stored in RDBMS (MySQL, Oracle, etc.) at one time, using Sqoop. In the realtime data gathering, data were gathered using Kafka. Sqoop designs import algorithms designed for gathering large-volume data, and designs export algorithms for loading analyzed data, while Kafka brings the DB updated at the Web servers in real time, using RESTful. The gathered data undergo diverse processes and are then loaded onto Hadoop. Future research needs to be furthered to develop the data gathering and storage tool, to analyze the stored data, to predict the driving environment, and to provide data fit for the users’ needs.