Conch: Competitive Debate Analysis via Visualizing Clash Points and Hierarchical Strategies

TL;DR

Conch introduces a novel, spiral-based visualization framework coupled with LLM-driven content extraction to analyze competitive debates. It jointly visualizes what is debated (clash points, disagreements, viewpoints) and how debates unfold (strategies, temporal evolution) across sessions. The approach is validated via two real-world case studies and a user study showing improved analysis speed and usable insights, with limitations addressed and future work outlined. This work advances debated content analysis by integrating hierarchical semantics with compact, multi-view visual analytics for coaches and debaters.

Abstract

In-depth analysis of competitive debates is essential for participants to develop argumentative skills and refine strategies, and further improve their debating performance. However, manual analysis of unstructured and unlabeled textual records of debating is time-consuming and ineffective, as it is challenging to reconstruct contextual semantics and track logical connections from raw data. To address this, we propose Conch, an interactive visualization system that systematically analyzes both what is debated and how it is debated. In particular, we propose a novel parallel spiral visualization that compactly traces the multidimensional evolution of clash points and participant interactions throughout debate process. In addition, we leverage large language models with well-designed prompts to automatically identify critical debate elements such as clash points, disagreements, viewpoints, and strategies, enabling participants to understand the debate context comprehensively. Finally, through two case studies on real-world debates and a carefully-designed user study, we demonstrate Conch's effectiveness and usability for competitive debate analysis.

Figures (10)

• Figure 1: The technical framework for segmenting blocks, extracting clash points, disagreements, and viewpoints, and constructing interaction paths.
• Figure 2: The glyph design of the Process View. (E) illustrates the chord diagram and its surrounding elements; (F) illustrates the structure of the sector-shaped areas; (F1) illustrates the disagreement block content.
• Figure 3: The glyph design of the Strategy View. (G) augmented stacked bar design, where column height encodes session length and column width encodes the peak usage frequency of strategies within each session; (H) lines and boxes design, illustrating strategy co-occurrence frequency.
• Figure 4: Alternative designs for the Strategy View.
• Figure 5: (a) Chordal diagram from Case Study 1. (b) S1: The disagreement and interaction path of the clash point ("Decision-Making Approach"); S2: Strategies View from Case Study 1.