Event relatedness assessment of Twitter messages for emergency response
作者：
Highlights：
• A novel method to assess the degree of relatedness of Twitter messages to a specific event of interest and identify the event-related tweets in a timely and accurate manner.
• Defining the concept of “event related term-classes” that are utilised as a template to identify and detect the tweets which are potentially related to the targeted event.
• Effective filtering and data cleaning techniques to facilitate and accelerate Twitter data analysis.
• An evaluation of the method in terms of accuracy, timeliness and reliability.
摘要
•A novel method to assess the degree of relatedness of Twitter messages to a specific event of interest and identify the event-related tweets in a timely and accurate manner.•Defining the concept of “event related term-classes” that are utilised as a template to identify and detect the tweets which are potentially related to the targeted event.•Effective filtering and data cleaning techniques to facilitate and accelerate Twitter data analysis.•An evaluation of the method in terms of accuracy, timeliness and reliability.
论文关键词：Emergency response,Social media,Twitter,Event detection
论文评审过程：Received 25 November 2015, Revised 31 August 2016, Accepted 6 September 2016, Available online 15 September 2016, Version of Record 23 November 2016.
论文官网地址：https://doi.org/10.1016/j.ipm.2016.09.002
