SNSJam: Road traffic analysis and prediction by fusing data from multiple social networks
作者：
Highlights：
• We propose SNSJam, which is a system to detect and predict road traffic jams by leveraging multiple data sources, specifically Twitter and Instagram.
• SNSJam supports multiple languages, specifically Arabic and English. It also supports Standard Arabic and UAE local Dialect.
• We developed a location recognizer that identifies locations from the text of posts and/or GPS locations. SNSJam supports user-defined locations, which are common names among people but different from the official names. SNSJam is the first such system to define and support user-defined locations.
• We developed a context-aware classifier to detect traffic jams. The classifier is able to identify the cause of traffic jams. The detected traffic jams can be visualized through a dynamic map.
• SNSJam employs a linear regression model to predict future traffic jams by leveraging current and historical posts.
摘要
•We propose SNSJam, which is a system to detect and predict road traffic jams by leveraging multiple data sources, specifically Twitter and Instagram.•SNSJam supports multiple languages, specifically Arabic and English. It also supports Standard Arabic and UAE local Dialect.•We developed a location recognizer that identifies locations from the text of posts and/or GPS locations. SNSJam supports user-defined locations, which are common names among people but different from the official names. SNSJam is the first such system to define and support user-defined locations.•We developed a context-aware classifier to detect traffic jams. The classifier is able to identify the cause of traffic jams. The detected traffic jams can be visualized through a dynamic map.•SNSJam employs a linear regression model to predict future traffic jams by leveraging current and historical posts.
论文关键词：Road traffic analysis,Traffic jam prediction,Data fusion,Data mining,Social networks
论文评审过程：Received 26 June 2019, Revised 28 September 2019, Accepted 4 October 2019, Available online 18 October 2019, Version of Record 18 October 2019.
论文官网地址：https://doi.org/10.1016/j.ipm.2019.102139
