Abstract

Mobile devices and Internet-based applications are producing a significant volume of data that may be used to, at least partially, replace some of the hardware necessary to sense traffic systems. However, there are several issues related to such an agenda: data are heterogeneous, unstructured, may appear in natural language, are normally not geolocated, and there are balancing issues related to the use of such data. This means that all these issues must be treated via software, especially using machine learning techniques. In this paper, a methodology is proposed, which is based on: extraction and processing of relevant information from social media; determination of its context; explanation of transportation related phenomena in terms of their contexts; and prediction of traffic conditions. The methodology was applied to a case study using data from the city of Porto Alegre, Brazil. Results shown that it was possible to associate traffic-related and context data to predict the traffic conditions that were originally reported in a Twitter account.

摘要

移动设备和基于 Internet 的应用程序正在产生大量数据，这些数据至少可以部分替代感知交通系统所需的一些硬件。然而，有几个与这样的议程相关的问题：数据是异构的、非结构化的、可能以自然语言出现、通常没有地理定位，并且存在与使用此类数据相关的平衡问题。这意味着所有这些问题都必须通过软件来处理，尤其是使用机器学习技术。本文提出了一种方法论，它基于：从社交媒体中提取和处理相关信息；确定其上下文；根据上下文解释交通相关现象；和预测交通状况。该方法用于使用巴西阿雷格里港市数据的案例研究。结果表明，可以将与交通相关的数据和上下文数据相关联，以预测最初在 Twitter 帐户中报告的交通状况。