Automatic detection of user trajectories from social media posts
作者：
Highlights：
• An automatic method for discovering user mobility patterns from social media posts.
• A novel text mining algorithm to automatically extract the main keywords in an area.
• A novel clustering algorithm to detect Regions-of-Interest from geotagged posts.
• A way to evaluate the accuracy of the different phases that compose our method.
• Experiments have been performed on a real dataset containing 3 million of Flickr items.
摘要
•An automatic method for discovering user mobility patterns from social media posts.•A novel text mining algorithm to automatically extract the main keywords in an area.•A novel clustering algorithm to detect Regions-of-Interest from geotagged posts.•A way to evaluate the accuracy of the different phases that compose our method.•Experiments have been performed on a real dataset containing 3 million of Flickr items.
论文关键词：Trajectory mining,RoI mining,Social media analysis,Geospatial clustering,Keyword extraction
论文评审过程：Received 3 December 2020, Revised 12 July 2021, Accepted 4 August 2021, Available online 12 August 2021, Version of Record 13 September 2021.
论文官网地址：https://doi.org/10.1016/j.eswa.2021.115733
