What is it about?
Recently, social media platforms, such as Twitter, have become important resources for users to report, seek and share information. During crisis events, messages posted on social media contain valuable information for situational updates and aid support. However, crisis-related posts are normally immersed in a high volume of irrelevant information. In this work, we focus on developing methods for classifying and summarizing tweets during crisis events. Our classification model is able to classify tweets into different humanitarian classes, such as infrastructure damage, caution and advice, injuries and death, rescue and donation, etc. Our summarizer generates a short summary of class-level tweets in near real-time.
Featured Image
Photo by Souvik Banerjee on Unsplash
Why is it important?
Our work helps users and human organizations to quickly catch up with the crisis situations, and avoid being overwhelmed by massive messages on social media. It can therefore serve as a tool for quick updates and aid support in case of crisis events.
Perspectives
The topic of this work has been becoming more and more attractive to the community. I hope our study can contribute some value to social media users, especially human organizations to better support people in need during crises.
Thi Huyen Nguyen
L3S Research Center, Hannover, Germany
Read the Original
This page is a summary of: Towards an Interpretable Approach to Classify and Summarize Crisis Events from Microblogs, April 2022, ACM (Association for Computing Machinery),
DOI: 10.1145/3485447.3512259.
You can read the full text:
Contributors
The following have contributed to this page
