1. Proceedings
  2. I3M
  3. 2022
  4. EMSS
  5. 10.46354/i3m.2022.emss.013

Recent trends in using machine learning and Twitter data for disaster management

  • a Maria Drakaki ,
  • b Vasileios Linardos,
  • c Panagiotis Tzionas
  • a,b International Hellenic University, Department of Science and Technology, University Center of International
    Programmes of Studies, 14th km Thessaloniki–N. Moudania, Thermi, 57001, Greece
  • c International Hellenic University Department of Industrial Engineering and Management, P.O. Box 141,
    Thessaloniki, 57400, Greece
Cite as
Drakaki M., Linardos V., and Tzionas P. (2022).,Recent trends in using machine learning and Twitter data for disaster management. Proceedings of the 34th European Modeling & Simulation Symposium (EMSS 2022). , 013 . DOI: https://doi.org/10.46354/i3m.2022.emss.013

Abstract

The impacts of disasters are devastating on the local communities and ecosystems. Recent advances in machine learning (ML)  including deep learning (DL) can cope with the complexity of disasters and have been used to develop methods that provide  effective solutions in all phases of disaster management. Furthermore; social media plays a critical role in communicating  disaster related information before and after a disaster strikes. ML and DL are being increasingly used to mine information from  social media data and especially Twitter data. This paper aims to provide the recent trends in this field by focusing on recent research studies presenting ML and DL based methods that leverage information posted on Twitter for disaster management.
The identified methods have been developed to provide solutions in the areas of disaster detection, damage assessment and post-disaster response.

References

  1. Alam, F.,Imran, M.,Ofli, F.(2017).Image4Act: Onlinesocial media image processing for disasterresponse. 2017 IEEE/ACM International Conferenceon Advances in Social Networks Analysis andMining 2017, Sydney, Australia,2017.
  2. Alexander, D.E.(2002).Principles of EmergencyPlanning and Management; Oxford UniversityPress, Oxford, UK, 2002.
  3. Arinta, RR.,Andi, EWR.(2019).Natural disasterapplication on big data and machine learning: Areview. 2019 4th International Conference onInformation Technology, Information Systems and Electrical Engineering (ICITISEE), Yogyakarta,Indonesia, 20-21 Nov. 2019.
  4. CRED & UNDRR.(2021).Global trends and perspectivesExecutive Summary. Doc Publ. Published online2021:8.
  5. Basu, M.,Shandilya, A.,Khosla, P.,Ghosh, K.,Ghosh,S.(2019).Extracting Resource Needs andAvailabilities from Microblogs for Aiding Post-Disaster Relief Operations.IEEE Trans Comput SocSyst., 6(3):604-618.
  6. Chaudhuri, N.,Bose, I.(2020).Exploring the role ofdeep neural networks for post-disaster decisionsupport. Decis Support Syst., 130:113234.
  7. Linardos, V., Drakaki, M., Tzionas, P., Karnavas, Y.L.(2022). Machine Learning in Disaster Management:Recent Developments in Methods and Applications.Mach. Learn.Knowl. Extr., 4(2):446-473.
  8. Holzinger, A.,Dehmer, M.,Emmert-Streib, F.,Cucchiara, R.,Augenstein, I.,Del Ser, J.,Samek, W.Jurisica, I.,Díaz-Rodríguez, N.(2022).Informationfusion as an integrativecross-cutting enabler toachieve robust, explainable, and trustworthymedical artificial intelligence.Inf. Fusion, 79: 263–278.
  9. Huang, X.,Li, Z.,Wang, C.,Ning, H.(2020).Identifying disaster related social media for rapidresponse: a visual-textual fused CNN architecture.Int J Digit Earth., 13(9):1017-1039.
  10. Kundu, S.,Srijith, PK.,Desarkar, MS.(2018).Classification of short-texts generated duringdisasters: A deep neural network based approach.2018 IEEE/ACM International Conference onAdvances inSocial Networks Analysis and Mining(ASONAM), Barcelona, Spain, 28-31 Aug. 2018.
  11. Layek, AK.,Poddar, S.,Mandal, S.(2019).Detection offlood images posted on online social media fordisaster response. 2019 Second InternationalConference on Advanced Computational andCommunication Paradigms (ICACCP).Gangtok,India, 25-28 Feb. 2019.
  12. Lecun, Y.,Bengio, Y.,Hinton, G.(2015).Deep learning.Nature. 521:436-444. doi:10.1038/nature14539.
  13. Li, H.,Caragea, D.,Caragea, C.,Herndon, N.(2018).Disaster response aided by tweet classification witha domainadaptation approach.J Contingencies CrisManag.,26(1):16-27.
  14. Li, T.,Li, Z.,Zhao, W.,et al.(2020).Analysis of medicalrescue strategies based on a rough set and geneticalgorithm: A disaster classification perspective. IntJ Disaster Risk Reduct., 42:101325.
  15. Nguyen, D.T.,Ofli, F.,Imran, M.,Mitra, P.(2017).Damage assessment from social media imagerydata during disasters. IEEE/ACM InternationalConference on Advances in Social NetworksAnalysis and Mining 2017, Sydney, Australia, 2017.
  16. Resch, B.,Usländer, F.,Havas, C.(2018).Combiningmachine-learning topic models andspatiotemporal analysis of social media data fordisaster footprint and damage assessment.CartogrGeogr Inf Sci.,45(4):362-376.
  17. Robertson, B.W.,Johnson, M.,Murthy, D.,Smith, WR.,Stephens, K.K.(2019).Using a combination ofhuman insights and ‘deep learning’ for real-timedisaster communication.Prog Disaster Sci., 2:
  18. O’Neal, A.,Rodgers, B.,Segler, J.,et al.(2018).Training anEmergency-Response Image Classifieron Signal Data. 2018 17th IEEE InternationalConference on Machine Learning and Applications(ICMLA), Orlando, FL, USA, 17-20 Dec. 2018.
  19. Reynard, D.,Shirgaokar, M.(2019).Harnessing thepower of machine learning: Can Twitter data beuseful in guiding resource allocation decisionsduring a natural disaster? Transp Res Part D TranspEnviron. 2019, 77:449-463.
  20. Sit, MA.,Koylu, C.,Demir, I.(2019).Identifyingdisaster-related tweets and their semantic, spatialand temporalcontext using deep learning, naturallanguage processing and spatial analysis: a casestudy of Hurricane Irma. Int J Digit Earth.,12(11):1205-1229.
  21. Sun, W..,Bocchini, P.,Davison, BD.(2020).Applications of Artificial Intelligence for DisasterManagement.Nat Haz, 103:2631-2689.
  22. UNISDR. Available online:https://www.undrr.org/publication/2020-non-covid-year-disasters.UNISDR, 2009. Availableonline:https://www.unisdr.org/files/7817_UNISDRTerminologyEnglish.pdf.
  23. Varga, I., Sano, M., Torisawa, K. Hashimoto, C.,Ohtake, K., Kawai, T. et al. (2013). Aid is out there:looking for help from Tweets during a large scaledisaster. Proceedings of the 51st Annual Meeting ofthe Association for Computational Linguistics,Curran Associates, Inc., Red Hook, NY, 1619-1629
  24. Wang, T.,Tao, Y.,Chen, SC.,Shyu ML.(2020).Multi-Task Multimodal Learning for Disaster SituationAssessment. 2020 IEEE Conference on MultimediaInformation Processing and Retrieval (MIPR),Shenzhen, China, 6-8 Aug. 2020.
  25. Yu, M.,Yang, C.,Li, Y.(2018).Big data in naturaldisaster management: A review.Geosci., 8.
  26. Zhang, DY.,Zhang, Y.,Li, Q.,Plummer, T.,Wang, D.(2019).CrowdLearn: A Crowd-AI hybrid system fordeep learning-based damage assessmentapplications. 2019 IEEE 39th InternationalConference on Distributed Computing Systems(ICDCS), Dallas, TX, USA, 7-10 July 2019.