Traffic observatory: a system to detect and locate traffic events and conditions using Twitter
Traffic observatory: a system to detect and locate traffic events and conditions using Twitter
复制标题
DOI：
10.1145/2442796.2442800
复制
发表时间：
2012-11
期刊：
影响因子：
--
通讯作者：
S. Ribeiro;C. Davis;Diogo Rennó Rocha de Oliveira;Wagner Meira Jr;Tatiana S. Gonçalves;G. Pappa
中科院分区：
文献类型：
--
作者：
S. Ribeiro;C. Davis;Diogo Rennó Rocha de Oliveira;Wagner Meira Jr;Tatiana S. Gonçalves;G. Pappa
Twitter 已成为共享用户生成内容的最受欢迎的平台之一，这些内容从普通对话到有关最近事件的信息。研究已经表明，推文的内容与现实世界中发生的事情具有高度的相关性。 Twitter 中经常谈论的一类事件是流量。为了帮助其他司机，许多用户在推特上发布有关当前交通状况的信息，甚至还有专门针对该主题的用户帐户。考虑到这一点，本文提出了一种方法来识别 Twitter 中的流量事件和状况，对其进行地理编码，并将其实时显示在 Web 上。初步结果表明，该方法能够以 50% 到 90% 的精度检测社区和主干道，具体取决于推文中提到的地点数量。
Twitter has become one of the most popular platforms for sharing user-generated content, which varies from ordinary conversations to information about recent events. Studies have already showed that the content of tweets has a high degree of correlation with what is going on in the real world. A type of event which is commonly talked about in Twitter is traffic. Aiming to help other drivers, many users tweet about current traffic conditions, and there are even user accounts specialized on the subject. With this in mind, this paper proposes a method to identify traffic events and conditions in Twitter, geocode them, and display them on the Web in real time. Preliminary results showed that the method is able to detect neighborhoods and thoroughfares with a precision that varies from 50 to 90%, depending on the number of places mentioned in the tweets.