GRAPE-MARS: A digital tool for enhanced multimodal discourse analysis in video-based research

Multimodal analyses play a crucial role in understanding meaning-making processes. However, their complexity and labor-intensive nature pose significant challenges. This paper introduces GRAPE-MARS (Multimodal Analysis Research Software), a web-based, open-source tool developed to enhance multimodal research by streamlining annotation, data retrieval, and visualization. GRAPE-MARS integrates AI-driven automatic transcription, user-friendly annotation, and advanced data analysis. In this paper, we present the design and functionalities of the software. Furthermore, we apply it in a case study to analyse engagement strategies in a YouTube science dissemination video. Using GRAPE-MARS, we quantify the distribution of semiotic modes and examine how engagement strategies are realized through distinct modal configurations. These results highlight the potential of the software to facilitate empirical multimodal analyses by providing structured, quantitative insights into semiotic interactions. This research contributes to the advancement of multimodal discourse analysis methodologies and underscores the value of digital tools in optimizing multimodal data coding and analysis.