论文标题
基于网络的北美城市公共交通系统分析
Network-Based Analysis of Public Transportation Systems in North American Cities
论文作者
论文摘要
实施了全面的数据分析系统，以提取信息和北美公共交通系统的比较。该系统基于传输系统的网络表示形式，并利用从图理论中已建立的属性到复杂域特定测量的指标和算法的跨度。由于大数据系统的性质和可伸缩性的要求，该系统中已经考虑了许多启发式优化和近似值。系统中还执行了与其他数据源的集成。对系统的子组件进行正式评估，以确保近似值具有合理的精度。旧金山，波士顿，多伦多和洛杉矶的四个城市进行比较的结果，批准了比较公共交通系统的大数据方法可以成功揭示基本的相似性和差异。
A comprehensive data analysis system is implemented for the extraction of information and comparison of North American public transport systems. The system is based on network representations of the transport systems and makes use of a span of metrics and algorithms from the established properties in graph theory to complicated domain specific measurements. Due to nature of big data systems and the requirement of scalability, many heuristic optimizations and approximations have been considered in the system. Integration with other sources of data specially population density maps is also executed in the system. Formal evaluations are done on subcomponents of the system to make sure the approximations have reasonable precision. Results on comparison of four cities, San Francisco, Boston, Toronto and Los Angeles, approves that the big data approach to comparison of public transit systems can successfully reveal the underlying similarities and differences.
