cs.HCMar 2, 2026

A Directed Graph Model and Experimental Framework for Design and Study of Time-Dependent Text Visualisation

Authors: Songhai Fan, Simon Angus, Tim Dwyer, Ying Yang, Sarah Goodwin, Helen Purchase

Organizations: Monash University

Read PDF
Ask AI
Cite

Copy citation

DOINot available for this paper

Abstract

Exponential growth in the quantity of digital news, social media, and other textual sources makes it difficult for humans to keep up with rapidly evolving narratives about world events. Various visualisation techniques have been touted to help people to understand such discourse by exposing relationships between texts (such as news articles) as topics and themes evolve over time. Arguably, the understandability of such visualisations hinges on the assumption that people will be able to easily interpret the relationships in such visual network structures. To test this assumption, we begin by defining an abstract model of time-dependent text visualisation based on directed graph structures. From this model we distill motifs that capture the set of possible ways that texts can be linked across changes in time. We also develop a controlled synthetic text generation methodology that leverages the power of modern LLMs to create fictional, yet structured sets of time-dependent texts that fit each of our patterns. Therefore, we create a clean user study environment (n=30) for participants to identify patterns that best represent a given set of synthetic articles. We find that it is a challenging task for the user to identify and recover the predefined motif. We analyse qualitative data to map an unexpectedly rich variety of user rationales when divergences from expected interpretation occur. A deeper analysis also points to unexpected complexities inherent in the formation of synthetic datasets with LLMs that undermine the study control in some cases. Furthermore, analysis of individual decision-making in our study hints at a future where text discourse visualisation may need to dispense with a one-size-fits-all approach and, instead, should be more adaptable to the specific user who is exploring the visualisation in front of them.

Figures & tables

NTVsNTVsDomainStory Context ElementNarrative Data TopologyEvaluation
  EntityEventTopicThreadTreeMapCase StudyUser StudyQl/Qt
1 ThemeRiver [ 9 ]News, Speeches  ✓✓  ✓✓Ql
2 TextFlow [ 5 ]Publications, News  ✓✓  ✓ Ql
3 CloudLines [ 14 ]News ✓ ✓  ✓ Ql
4 EventRiver [ 20 ]News ✓✓✓  ✓✓Ql/Qt
5 Metro Map [ 28 , 30 ]News ✓✓  ✓ ✓Ql
Table 1: Taxonomy of Narrative Visualisation Techniques: This table categorises various time-dependent text visualisation techniques based on three dimensions: the domain of text data (e.g., news), Story Context Elements (entity, event, topic), Narrative Data Topologies (thread, tree, map), and evaluation methods (whether a case study or user study is involved, and whether the approach is qualitative (Ql) or quantitative (Qt)). Row numbers correspond to labelled example images in A Directed Graph Model and Experimental Framework for Design and Study of Time-Dependent Text Visualisation . Ticks indicate the presence of specific characteristics within each technique.
Figure 1: Representative narrative structures identified in our literature survey. They have in common that one spatial dimension indicates time (here, left-to-right). Different structures afford different ways to show relationships between announcements. While all use linear elements to connect pairs of announcements, the non-time dimension (here vertical) can show secondary relationships (such as shared events in storyline views) or strict thematic separation (such as thread views). Note that River is a subtype of Thread that additionally encodes time-dependent quantity (typically term frequency).
Figure 2: Left: An illustration of a Time-Track Narrative Graph Model. Right: The three node motifs for the Time-Track Narrative Graph Model include Sequential motifs (Linear, Arch, Ladder, Early Turn, Late Turn) and Non-Sequential motifs (Sharp Branch, Wide Branch, Sharp Merge, Wide Merge).
Figure 3: The Graph-to-Text Pipeline . This diagram illustrates the process of converting a TTNG into synthetic news text through three main stages: (1) Enriching the graph with narrative context using the Crafter; (2) Mapping narrative context elements with the Cartographer; (3) Generating the final synthetic news text with the Writer. The process enforces structural patterns through shared attributes while allowing natural thematic connections to emerge. For additional examples of narrative context, prompt templates, and the map algorithm, refer to the supplementary materials A .
Figure 4: Distribution of similarity scores for Announcement pairs in our synthetic dataset. Box extents give 25th to 75th percentiles, vertical line gives median, whilst extent of horixontal lines indicate min–max. Red boxes show similarity scores for Announcements from different tracks (expected to be lower), while blue boxes show scores for Announcements in the same track (expected to be higher). Across all three metrics (Jaccard, TF-IDF, and BERT), there is clear separation between same-track and different-track pairs, with same-track pairs consistently showing higher similarity scores compared to different-track pairs.
Figure 5: Confusion matrix comparing participant-selected motifs (rows) against predefined motifs (columns). Diagonal cells indicate correct identifications. The Grand Total column shows total selections per motif. Each column sums to 30, controlled by the graph-to-text pipeline input. Numbered talk bubbles correspond to detailed analysis in 6.2 .
Figure 6: Confusion case study for LateTurn
Figure 7: Confusion case study for EarlyTurn
Figure 8: Distribution of reasoning categories for narrative selections. Top bars show distribution when participants correctly identified motifs, Bottom bars show distribution for incorrect identifications.
Appendix figures & tables3 assets

Supplementary material from the paper’s appendix.

Appendix

Figure 9: Example story graph structure
Figure 10: Two TTNG views for the same Little Red Riding Hood statements. Top: Topic-organised view ( α=Topic\alpha=Topic ), showing Journey →\rightarrow Deception →\rightarrow Rescue. Bottom: Entity-organised view ( α=Entity\alpha=Entity ), emphasising character interactions.
IDEntitiesTopicFactual statement
V1RedJourneyRed departs her home for Grandma’s cottage carrying a basket of food through the woods.
V2WolfJourneyThe Wolf watches Red from the bushes near the dark forest entrance to plan his hunt.
V3Red, WolfJourneyRed stops to chat with the Wolf on the wooded path because he seems friendly.
V4Wolf, RedDeceptionThe Wolf encourages Red to pick wild flowers to delay the girl from reaching the cottage.
V5WolfDeceptionThe Wolf eats Grandma and dons her nightcap at the cottage to trick the girl.
V6Hunter, WolfRescueThe Hunter tracks the Wolf across the forest floor after hearing a suspicious disturbance.
Appendix
Table 2: Factual statements and attributes used in the Little Red Riding Hood TTNG example.
More in this topic

Data Visualization

Data visualization studies how visual representations help people explore, interpret, communicate, and reason about data.

See all 120 papers

Explore similar work

CardsList

  1. True (VIS) Lies: Analyzing How Generative AI Recognizes Intentionality, Rhetoric, and Misleadingness in Visualization Lies

    Apr 1, 2026Graziano Blasilli, Marco AngeliniData VisualizationVLM Evaluation

    Sapienza University of Rome, Rome, Italy. · Link Campus University, Rome, Italy.

  2. VisInteract: Towards Dynamic Interactive Text-to-Visualization under Imperfect Queries

    Sep 14, 2026Wenxin Xu, Jinwei Lu, Hwanhee Kim +4Data VisualizationMonte Carlo Tree Search

    The Hong Kong Polytechnic University · ByteDance

  3. Evaluating Reasoning Fidelity in Visual Text Generation

    Jun 3, 2026Jiajun Hong, Jiawei ZhouT2I Generation EvaluationText Rendering in T2I Generation

    Stony Brook University