SpreadLine: Visualizing Egocentric Dynamic Influence

TL;DR

SpreadLine addresses the challenge of visualizing dynamic egocentric networks across four interrelated aspects—strength, function, structure, and content—by introducing a storyline-based framework that encodes these dimensions at the microscopic level using a metro-map-inspired layout. Building on a taxonomy of egocentric-network analysis tasks, SpreadLine provides customizable encodings, including space division, block distinctions, line colors, and a contextual affinity view to unify temporal and contextual information. The framework is demonstrated through three real-world case studies (disease propagation, social media trends, and academic career evolution) and a usability study, highlighting its generalizability, interpretability, and potential limitations in scalability. The work contributes a task-driven design, a flexible visualization approach, and practical guidelines for tailoring SpreadLine to diverse analytic goals, with future directions in onboarding, design space formalization, and data-story authoring tools.

Abstract

Egocentric networks, often visualized as node-link diagrams, portray the complex relationship (link) dynamics between an entity (node) and others. However, common analytics tasks are multifaceted, encompassing interactions among four key aspects: strength, function, structure, and content. Current node-link visualization designs may fall short, focusing narrowly on certain aspects and neglecting the holistic, dynamic nature of egocentric networks. To bridge this gap, we introduce SpreadLine, a novel visualization framework designed to enable the visual exploration of egocentric networks from these four aspects at the microscopic level. Leveraging the intuitive appeal of storyline visualizations, SpreadLine adopts a storyline-based design to represent entities and their evolving relationships. We further encode essential topological information in the layout and condense the contextual information in a metro map metaphor, allowing for a more engaging and effective way to explore temporal and attribute-based information. To guide our work, with a thorough review of pertinent literature, we have distilled a task taxonomy that addresses the analytical needs specific to egocentric network exploration. Acknowledging the diverse analytical requirements of users, SpreadLine offers customizable encodings to enable users to tailor the framework for their tasks. We demonstrate the efficacy and general applicability of SpreadLine through three diverse real-world case studies (disease surveillance, social media trends, and academic career evolution) and a usability study.

Figures (5)

• Figure 1: A foundational configuration of SpreadLine. Left: The node-link representation of the trade network among 6 animal farms across two timepoints. Right: Using the same information depicted on the left, SpreadLine employs various encodings to condense crucial network information, emphasizing the function, structure, and strength aspects.
• Figure 2: A disease outbreak detected on Farm SI (ego) on March 13th. To identify farms at risk of infection, an animal health specialist can investigate timepoints of interest (1, 2, and 3) and examine the farms from trading relationships, health conditions, and geospatial proximity (5, 6, and 8) through user interactions (4 and 7). The ego is indicated as a larger point in the expanded view. The layout is computed with vertical space optimization.
• Figure 3: Public reaction to the sentence of the actor Danny Masterson. (A) The overview. Based on the public engagement, reflected in block distinction, we identify three distinct periods (P1, P2, and P3). (B) The focused examination on common topics. The slider is repositioned for presentation. The following are knowledge graphs of tweets on different days (entities are annotated for presentation): (C) On September 7th, opposing sides discussed evidence associated with the sentence. (D) On September 10th, the supporting sides condemned the two supporters of the ego and refused their apology. (E) On September 16th, there was a resurgence of public reaction due to the involvement of another individual in the #MeToo movement. Note that vertical space optimization is employed with entity stacking.
• Figure 4: Career evolution of two researchers. Top: Tamara Munzner with 123 co-authors from 1994 to 2022. Bottom: Jeffrey Heer with 198 co-authors from 2002 to 2022. (1)(2) Research focus of Tamara in 2003 and 2012, respectively. Straight line optimization is employed with entity stacking.
• Figure 5: Histograms of participants' ratings (7-point Likert scale) on the overall utility and usability of SpreadLine.