Lending Ears to Unheard Voices: An Empirical Analysis of User-Generated Content on Social Media During a Humanitarian Crisis

Governments and healthcare organizations increasingly pay attention to social media and big data for handling a disease outbreak. The institutions and organizations need information support for gaining insights into the situation and act accordingly. However, in the field of humanitarian and healthcare operations, the research on the use of social media and big data for providing information support is limited. This study attempts to fill the above-mentioned gap by investigating the relationship between the activity on social media and the quantum of the outbreak, and further use content analytics to construct a model for segregating incoming feeds. We use the case example of the COVID-19 outbreak. The findings show that social media activity is a reflection of the outbreak situation on the ground. In particular, we find that tweets posted by people during a crisis outbreak are indeed an indicator of the rise in cases of the disease. Further, we find that the presence of negative tweets and an increase in sharing of tweets explain the rise in cases of a disease outbreak. Our results also show that negative tweets and lengthy tweets with mixed emotions lead to more diffusion of information. Building further on this insight, we propose a model using advanced analytical methods to reduce a large amount of unstructured data to four key categories in real-time. These categories are––irrelevant posts, emotional outbursts, distress alarm, and relief measures. The supply-side stakeholders (such as policymakers and humanitarian organizations) could use this information in real-time and optimize resources and relief packages in the right direction.