A Big Data Architecture for Near Real-time Traffic Analytics
| Autor: | Yikai Gong, Paul Rimba, Richard O. Sinnott |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
0106 biological sciences
Data processing Traffic analysis business.industry Computer science Sydney Coordinated Adaptive Traffic System Real-time computing Big data Cloud computing 02 engineering and technology 01 natural sciences 010601 ecology Identification (information) Analytics 0202 electrical engineering electronic engineering information engineering 020201 artificial intelligence & image processing business Spatial analysis |
| Zdroj: | UCC (Companion") |
| DOI: | 10.1145/3147234.3151010 |
| Popis: | Big data is a popular research topic that has brought about a range of new IT challenges and opportunities. The transport domain is one area that has much to benefit from big data platforms. It requires capabilities for processing voluminous amounts of heterogeneous data that is often created in near real time and at high velocity from a multitude of distributed sensors. It can also require the application of performance-oriented spatial data processing of such data. In this paper, we present a platform (SMASH) that tackles many of the specific challenges raised by the transport domain. We present a range of case studies applying SMASH to transport and other data used to understand traffic phenomenon across the State of Victoria, Australia. The novelty of this work is that this Cloud-based platform is not designed for a specific type of data or for a specific form of data processing. Rather it supports a range of data flavours with a range of data processing possibilities. In particular we show how the platform can be used for analyzing social media data used for traffic jam identification through spatial and temporal clustering tweets on the road network and compare the results with official real-time traffic data based on the Sydney Coordinated Adaptive Traffic System (SCATS - www.scats.com.au) that has been rolled out across Victoria. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|