In transportation, ubiquitous deployment of low-cost sensors combined with powerful computer hardware and high-speed network makes big data available. USDOT defines big data research in transportation as a number of advanced techniques applied to the capture, management and analysis of very large and diverse volumes of data. Data in transportation are usually well organized into tables and are characterized by relatively low dimensionality and yet huge numbers of records. Therefore, big data research in transportation has unique challenges on how to effectively process huge amounts of data records and data streams. The purpose of this study is to conduct research on the problems caused by large data volume and data streams and to develop applications for data analysis in transportation. To process large number of records efficiently, we have proposed to aggregate the data at multiple resolutions and to explore the data at various resolutions to balance between accuracy and speed. Techniques and algorithms in statistical analysis and data visualization have been developed for efficient data analytics using multiresolution data aggregation. Results will be helpful in setting up a primitive stage towards a rigorous framework for general analytical processing of big data in transportation.
WMU ScholarWorks Citation
Yang, Li; Cho, Hyunkeun; and Oh, Jun-Seok, "14-08 Big Data Analytics to Aid Developing Livable Communities" (2015). Transportation Research Center Reports. 32.