SAT-Geo: A social sensing based content-only approach to geolocating abnormal traffic events using syntax-based probabilistic learning
作者：
Highlights：
• Study the problem of geolocating abnormal traffic events using social media data.
• Develop a novel syntax-based learning approach to identify location entities.
• Design a distance-aware estimation method to estimate the geolocation.
• Evaluate the estimation performance of SAT-Geo on two real-world datasets.
• Evaluation results show significant performance gains of the proposed solution.
摘要
•Study the problem of geolocating abnormal traffic events using social media data.•Develop a novel syntax-based learning approach to identify location entities.•Design a distance-aware estimation method to estimate the geolocation.•Evaluate the estimation performance of SAT-Geo on two real-world datasets.•Evaluation results show significant performance gains of the proposed solution.
论文关键词：Syntax-based learning,Abnormal detection,Geolocalization,Social sensing
论文评审过程：Received 16 May 2021, Revised 21 September 2021, Accepted 24 October 2021, Available online 17 November 2021, Version of Record 17 November 2021.
论文官网地址：https://doi.org/10.1016/j.ipm.2021.102807
