Cross-Lingual Query-Based Summarization of Crisis-Related Social Media: An Abstractive Approach Using Transformers | Proceedings of the 33rd ACM Conference on Hypertext and Social Media
Authors:
Open Access
- 28 June 2022
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 21-31
Abstract
Relevant and timely information collected from social media during crises can be an invaluable resource for emergency management. However, extracting this information remains a challenging task, particularly when dealing with social media postings in multiple languages. This work proposes a cross-lingual method for retrieving and summarizing crisis-relevant information from social media postings. We describe a uniform way of expressing various information needs through structured queries and a way of creating summaries answering those information needs. The method is based on multilingual transformers embeddings. Queries are written in one of the languages supported by the embeddings, and the extracted sentences can be in any of the other languages supported. Abstractive summaries are created by transformers. The evaluation, done by crowdsourcing evaluators and emergency management experts, and carried out on collections extracted from Twitter during five large-scale disasters spanning ten languages, shows the flexibility of our approach. The generated summaries are regarded as more focused, structured, and coherent than existing state-of-the-art methods, and experts compare them favorably against summaries created by existing, state-of-the-art methods.Keywords
Funding Information
- Ministry of Science and Innovation of Spain (PID2019-109064GB-I00)
- La Caixa Foundation (100010434)
- Horizon 2020 (871042)
This publication has 26 references indexed in Scilit:
- Classifying and Summarizing Information from Microblogs During EpidemicsInformation Systems Frontiers, 2018
- Guiding Generation for Abstractive Text Summarization Based on Key Information Guide Network2018
- MULTILINGUAL DETECTION AND MAPPING OF EMERGENCY AND DISASTER-RELATED TWEETSMATTER: International Journal of Science and Technology, 2017
- Centroid-based Text Summarization through Compositionality of Word Embeddings2017
- Extracting Situational Information from Microblogs during Disaster Events2015
- Geographic Situational Awareness: Mining Tweets for Disaster Preparedness, Emergency Response, Impact, and RecoveryISPRS International Journal of Geo-Information, 2015
- Processing Social Media Messages in Mass EmergencyACM Computing Surveys, 2015
- The Case for Readability of Crisis Communications in Social Media2015
- A geographic approach for combining social media and authoritative data towards identifying useful information for disaster managementInternational Journal of Geographical Information Science, 2015
- Comparative Study of Text Summarization MethodsInternational Journal of Computer Applications, 2014