Harnessing Digital Data for Outbreak Management: A Generative Agent-Based Policy Formulation and Assessment

Aparup Khatua, Rui Mao, Erik Cambria

Keywords: Generative Agents, Crisis Management, Social Media Data, Policy Formulation, Policy Assessment, TinyTroupe

Abstract: Extant literature has rarely considered the cognitive aspects of an outbreak. Social psychology literature suggests that the mental construal of a specific event depends on psychological distance, i.e., the perceived distance from the event. Consequently, psychological distance can potentially influence (or even distort) the linguistic content of deliberations during outbreaks, making policy formulation based on such data problematic. Hence, we employed the theoretical lens of Construal-level Theory to understand the cognitive aspects of social media data during the 2014 Ebola outbreak and used these insights as input for our proposed generative agent-based policy formulation. The \textbf{core contribution} of this study is \textit{leveraging social media data for policymaking through generative agents} and assessing the same by comparing the suggested policies with the original policy documents.

Submission Type: Long Paper (archival)

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 11