DISCOVERING DISASTER EVENTS FROM SOCIAL MEDIA STREAMS

Authors

  • Yue-Fu Tsai National Cheng Kung University
  • Jih-Liang Hsieh National Cheng Kung University
  • Wei-Guang Teng National Cheng Kung University
  • Ting-Wei Hou National Cheng Kung University
  • Chih-Pin Freg National Cheng Kung University
  • Yu-Chung Tsao National Taiwan University of Science and Technology

DOI:

https://doi.org/10.23055/ijietap.2018.25.5.3701

Keywords:

data analytics, disaster management, event detection, social network analysis

Abstract

Natural and man-made disasters can both cause severe loss of lives and economic damages. Examples include earthquakes, floods, and road crashes. Nevertheless, to rapidly and accurately identify the latest status of a disaster event is undoubtedly one of the most difficult tasks for agencies in crisis management. In this work, we thus propose to monitor online data streams in social media for detecting and tracking real world events. Unlike conventional media, social media is advantageous because of its immediateness, huge data scale, and worldwide availability. Nevertheless, the messages generated by netizens could be incomplete, subjective, or even error prone. Only with an appropriately designated scheme, invaluable clues embedded in huge amounts of online messages can be discovered when carefully exploiting the information over content, temporal, and social dimensions. Specifically, we collect data from multiple social networks, conduct real-time analysis, and present interactive visualization. Experimental studies show that the proposed scheme is demonstrated to be feasible for agencies in practice.

Published

2019-01-03

How to Cite

Tsai, Y.-F., Hsieh, J.-L., Teng, W.-G., Hou, T.-W., Freg, C.-P., & Tsao, Y.-C. (2019). DISCOVERING DISASTER EVENTS FROM SOCIAL MEDIA STREAMS. International Journal of Industrial Engineering: Theory, Applications, and Practice, 25(5). https://doi.org/10.23055/ijietap.2018.25.5.3701