Leveraging LLMs for Persona-Based Visualization of Election Data
Abstract
Visualizations are essential tools for disseminating information regarding elections and their outcomes, potentially influencing public perceptions. Personas, delineating distinctive segments within the populace, furnish a valuable framework for comprehending the nuanced perspectives, requisites, and behaviors of diverse voter demographics. In this work, we propose making visualizations tailored to these personas to make election information easier to understand and more relevant. Using data from UK parliamentary elections and new developments in Large Language Models (LLMs), we create personas that encompass the diverse demographics, technological preferences, voting tendencies, and information consumption patterns observed among voters.Subsequently, we elucidate how these personas can inform the design of visualizations through specific design criteria. We then provide illustrative examples of visualization prototypes based on these criteria and evaluate these prototypes using these personas and LLMs. We finally propose some actionable insights based upon the framework and the different design artifacts.