Needling Through the Threads: A Visualization Tool for Navigating Threaded Online Discussions

TL;DR

Needle addresses the challenge of moderating large-scale threaded online discussions by providing a visualization-driven tool that summarizes activity, toxicity, and voting trends over time. The approach combines feed-level overviews with thread-level drill-down visualizations, enabling multi-level awareness and fast triage. The paper contributes a novel sociotechnical system, an evaluation with Reddit moderators (N=10), and design guidelines for visualization-based moderation tools. The findings show reduced cognitive load, improved prioritization, and support for human-in-the-loop decision-making, with implications for customization and proactive moderation.

Abstract

Navigating large-scale online discussions is difficult due to the rapid pace and large volume of user-generated content. Prior work in CSCW has shown that moderators often struggle to follow multiple simultaneous discussions, track evolving conversations, and maintain contextual understanding--all of which hinder timely and effective moderation. While platforms like Reddit use threaded structures to organize discourse, deeply nested threads can still obscure discussions and make it difficult to grasp the overall trajectory of conversations. In this paper, we present an interactive system called Needle to support better navigation and comprehension of complex discourse within threaded discussions. Needle uses visual analytics to summarize key conversational metrics--such as activity, toxicity levels, and voting trends--over time, offering both high-level insights and detailed breakdowns of discussion threads. Through a user study with ten Reddit moderators, we find that Needle supports moderation by reducing cognitive load in making sense of large discussion, helping prioritize areas that need attention, and providing decision-making supports. Based on our findings, we provide a set of design guidelines to inform future visualization-driven moderation tools and sociotechnical systems. To the best of our knowledge, Needle is one of the first systems to combine interactive visual analytics with human-in-the-loop moderation for threaded online discussions.

Figures (4)

• Figure 1: Threaded Discussion A is a mock discussion designed to reflect the scale and comment volume of a real large-scale Reddit thread. Threaded Discussion B illustrates the "threads" that occur at multiple levels within online discussions using synthetic data.
• Figure 2: Needle's System Architecture. Data from Reddit is processed and augmented with toxicity scores, stored as a JSON cache, and served by a Node.js backend to a Next.js/D3.js frontend; front-end is powered by two pages: Feed-level and Thread-level interfaces.
• Figure 3: The feed-level interface of Needle, which presents subreddit-wide comment activity and toxicity. (A) Control panel for setting time (a1), toxicity and score threshold (a2), zoom level for time range (a3), line toggles with legend (a4), and toggle show inactive posts (a5). (B) Global histograms for toxicity (b1) and score distributions (b2). (C) Sorting dropdown to organize posts by criteria. (D) Post summaries, with a segmented comment breakdown bar (d1), and a temporal activity chart (d2).
• Figure 4: The thread-level interface of Needle, which provides detailed visualization of individual discussion threads. (E) Control panel with the same customization options as the feed-level interface, adding two new features of applying highlights to comments (e1) and filter comments (e2). (F) Comment visualization features including: active TLCs only indicator (f1), hover effects to indicate linkage between visualizations to comments (f2), jump to Comment button for quick navigation (f3), and temporal activity chart for the selected TLC (f4). (G) Actual comments: toxic comment highlighting (g1), hover effects (g2), navigation controls to move between charts and comments (g3), and moderation buttons (g4).