Abstract
In the early stages of an emergency, information extracted from social media can support crisis response with evidence-based content. In order to capture this evidence, the events of interest must be first promptly detected. An automated detection system is able to activate other tasks, such as preemptive data processing for extracting event-related information. In this paper, we extend the human-in-the-loop approach in our previous work, TriggerCit, with a machine-learning-based event detection system trained on word count time series and coupled with an automated lexicon building algorithm. We design this framework in a language-agnostic fashion. In this way, the system can be deployed to any language without substantial effort. We evaluate the capacity of the proposed work against authoritative flood data for Nepal recorded over two years.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
Notes
- 1.
Searching on social media is usually done with keywords, in combination with logical and advanced operators.
- 2.
Following canonical machine learning terminology, we will henceforth refer to the estimations of an algorithm on real-time data as “predictions”.
- 3.
- 4.
We did not use previous reports since, ostensibly, the data collection process changed at some point.
- 5.
- 6.
- 7.
This is mandatory on Twitter since a query consisting of only stopwords is rejected.
- 8.
By mixing manual exploration and automatic exploration using KerasTuner.
- 9.
- 10.
- 11.
Approximately 5 million tweets sampled from positive and negative days.
- 12.
We did not use Global Flood Monitor data described in [4] since it does not contain certified data, and we were not able to obtain NatCatSERVICE data from Munich Re.
- 13.
The International Disaster database, https://www.emdat.be/.
- 14.
References
Autelitano, A., Pernici, B., Scalia, G.: Spatio-temporal mining of keywords for social media cross-social crawling of emergency events. GeoInformatica 23(3), 425–447 (2019)
Avvenuti, M., Cimino, M.G.C.A., Cresci, S., Marchetti, A., Tesconi, M.: A framework for detecting unfolding emergencies using humans as sensors. Springerplus 5, 43 (2016)
Bono, C., Pernici, B., Fernandez-Marquez, J.L., Shankar, A.R., Mülâyim, M.O., Nemni, E.: TriggerCit: early flood alerting using twitter and geolocation-a comparison with alternative sources. In: Proceedings of ISCRAM 2022, Tarbes, France (2022)
de Bruijn, J.A., de Moel, H., Jongman, B., de Ruiter, M.C., Wagemaker, J., Aerts, J.C.J.H.: A global database of historic and real-time flood events based on social media. Sci. Data 6(1), 311 (2019)
Havas, C., et al.: E2mc: improving emergency management service practice through social media and crowdsourcing analysis in near real time. Sensors 17(12), 2766 (2017)
Imran, M., Castillo, C., Diaz, F., Vieweg, S.: Processing social media messages in mass emergency: survey summary. In: Companion Proceedings of the The Web Conference WWW 2018, Lyon, France, pp. 507–511 (2018)
Ismail Fawaz, H., Forestier, G., Weber, J., Idoumghar, L., Muller, P.A.: Deep learning for time series classification: a review. Data Min. Knowl. Discovery 33(4), 917–963 (2019)
Jongman, B., Wagemaker, J., Romero, B.R., De Perez, E.C.: Early flood detection for rapid humanitarian response: harnessing near real-time satellite and twitter signals. ISPRS Int. J. Geo-Inf. 4(4), 2246–2266 (2015)
Kingma, D.P., Ba, J.: Adam: a method for stochastic optimization (2014). arxiv:1412.6980Comment. Published as a conference paper at the 3rd International Conference for Learning Representations, San Diego (2015)
Lindersson, S., Brandimarte, L., Mård, J., Di Baldassarre, G.: A review of freely accessible global datasets for the study of floods, droughts and their interactions with human societies. WIREs Water 7(3), e1424 (2020)
Lorini, V., Castillo, C., Nappo, D., Dottori, F., Salamon, P.: Social media alerts can improve, but not replace hydrological models for forecasting floods. In: 2020 IEEE/WIC/ACM International Joint Conference on Web Intelligence and Intelligent Agent Technology (WI-IAT), pp. 351–356. IEEE Computer Society (2020)
Munawar, H.S., Hammad, A.W.A., Waller, S.T.: Remote sensing methods for flood prediction: a review. Sensors 22(3), 960 (2022)
Nemni, E., Bullock, J., Belabbes, S., Bromley, L.: Fully convolutional neural network for rapid flood segmentation in synthetic aperture radar imagery. Remote Sens. 12(16), 2532 (2020)
Ofli, F.: A real-time system for detecting landslide reports on social media using artificial intelligence. In: Di Noia, T., Ko, I.Y., Schedl, M., Ardito, C. (eds.) Web Engineering, pp. 49–65. Springer International Publishing, Cham (2022)
Olteanu, A., Castillo, C., Diaz, F., Vieweg, S.: Crisislex: a lexicon for collecting and filtering microblogged communications in crises. Proc. Int. AAAI Conf. Web Soc. Media 8(1), 376–385 (2014)
Perera, D., et al.: Flood early warning systems: A review of benefits, challenges and prospects (2019)
Sakaki, T., Okazaki, M., Matsuo, Y.: Tweet analysis for real-time event detection and earthquake reporting system development. IEEE Trans. Knowl. Data Eng. 25(4), 919–931 (2013)
Shoyama, K., Cui, Q., Hanashimaa, M., Sano, H., Usuda, Y.: Emergency flood detection using multiple information sources: integrated analysis of natural hazard monitoring and social media data. Sci. Total Environ. 767(144371), 1–11 (2021)
Stollberg, B., De Groeve, T.: The use of social media within the global disaster alert and coordination system (GDACS). In: Proceedings of the 21st International Conference on World Wide Web, pp. 703–706 (2012)
United Nations Office for the Coordination of Humanitarian Affairs: Five essentials for the first 72 hours of disaster response. https://www.unocha.org/story/five-essentials-first-72-hours-disaster-response. Accessed 15 Aug 2022
Wang, Z., Yan, W., Oates, T.: Time series classification from scratch with deep neural networks: a strong baseline. In: 2017 International Joint Conference on Neural Networks (IJCNN), pp. 1578–1585 (2017)
Acknowledgements
The work at Politecnico di Milano and IIIA-CSIC was funded by the European Commission H2020 Project Crowd4SDG, #872944.
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 IFIP International Federation for Information Processing
About this paper
Cite this paper
Bono, C.A., Mülâyim, M.O., Pernici, B. (2023). Learning Early Detection of Emergencies from Word Usage Patterns on Social Media. In: Gjøsæter, T., Radianti, J., Murayama, Y. (eds) Information Technology in Disaster Risk Reduction. ITDRR 2022. IFIP Advances in Information and Communication Technology, vol 672. Springer, Cham. https://doi.org/10.1007/978-3-031-34207-3_20
Download citation
DOI: https://doi.org/10.1007/978-3-031-34207-3_20
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-031-34206-6
Online ISBN: 978-3-031-34207-3
eBook Packages: Computer ScienceComputer Science (R0)