DiscourseDash: designing and evaluating a tool for analysing social media discourses to support decision-making
Year: 2026
Type: article
Abstract: Social media discourses have become a significant source of information for decision-makers across various fields, including public health, policy, and brand management. Although analytical tools exist to support these practices, there is a limited understanding of how industry experts engage with and incorporate a discourse-level analysis perspective into their decision-making workflows. To address this gap, we developed DiscourseDash, a dashboard tool that integrates stance detection, sentiment analysis, harmfulness detection, and popularity analysis of evolving social media discourses. DiscourseDash provides a structured view of discourse-level dynamics to support users in interpreting not just what is happening, but why. DiscourseDash differentiates itself from existing tools that primarily visualise high-level statistics by emphasising transparency and collaborative decision-making. We conducted a two-phase study: (1) a mixed-method evaluation with general users to assess effectiveness in navigating social media discourses within a scenario-based decision-making compared to an unstructured method, and (2) in-depth interviews with industry experts to understand their workflows, gather perceptions of DiscourseDash, and explore alignment with real-world practices. Our findings underscore the value of qualitative understanding in social media analysis by demonstrating how discourse-level insights support interpretation and encourage strategic decision-making. In recognising the central role of human expertise, our work highlights the importance of collaborative design by facilitating human interpretation with the prototype. This article contributes practical insights through a conceptual and design-led approach to social media analysis for decision-making.
Source: PeerJ Computer Science
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $1,395
OpenAlex ID: W7130386868