The Employment of Modern Technologies in Analyzing the Stylistic Diversity of Graphic Discourse in Election Campaigns

This study examines the potential of modern technologies, such as machine learning and image analysis, to analyze the stylistic diversity of graphic discourse in election campaigns. As the complexity and volume of campaign materials have grown through the years, it has become clear that automated, objective methods of analyzing visual and textual content are needed. This research employs CNN and NLP models to categorize and assess campaign graphics based on design style, sentiment and color usage based on campaign context. Results show the models achieving a high sentiment positive score indicating that simple (minimalist and modern) design choices characterize campaign literature, where favorable textual sentiment predominates, and color is used effectively to signal political preferences. These findings offer new possibilities for the use of AI tools to create deeper understanding of how design influences political discourse, empowering campaigns with the ability to iterate on their messaging for greater voter reach and persuasion ability.