Pith. sign in

REVIEW 3 major objections 8 minor 72 references

Leveraging LLMs for Persona-Based Visualization of Election Data

T0 review · 3 major / 8 minor · reviewed 2026-08-06 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read LLM-generated voter personas can design and score election visualizations

desk verdict A coherent workflow sketch whose load-bearing persona statistics come from an unverified LLM deep-research report; useful as a cautionary example, not yet as a validated method. read the letter →

arxiv 2507.21900 v1 pith:O4GUKSPN submitted 2025-07-29 cs.HC

Swaroop Panda , Arun Kumar Sekar This is my paper
classification cs.HC
keywords electiondatavisualizationuserpersonasLLM-generateduser-centereddesigncriteriaprototypesheuristicevaluationUKelections
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper tries to show that LLM-generated personas can carry a user-centered design process for election data from start to finish: personas define who the visualizations are for, design criteria translate persona traits into concrete chart and medium choices, prototypes embody those choices, and LLM evaluation checks the result. It builds three UK voter personas and three corresponding prototypes—an infographic for print, a short data video for social media, and an interactive smartphone view—and evaluates all three with an LLM prompted to apply established visualization heuristics. If the claim holds, designers could produce audience-specific election visualizations and assess them without traditional user studies, which matters because election graphics shape how people understand results and because conventional persona research is slow and expensive. The paper also argues that personas personalize visualizations, LLMs can automate intermediate steps, and designers remain responsible for curation and direction.

What carries the argument

The carrying object is the persona as a structured intermediate representation. Each persona is defined by four attributes: a short description, technology attitudes, voting behavior, and information-consumption choices. The design rule is that technology attitudes select the dissemination medium, information-consumption choices select the visual artifact and visual design, and voting behavior selects the chart types and the data shown. A second piece of machinery is the LLM evaluation protocol: prototypes are handed to the LLM as images, the LLM first describes what it sees, then scores the design against published visualization heuristics and against the persona description, so the same generation tool also provides the assessment.

What would settle it

Retrieve the three survey sources cited in the appendix and check the quoted statistics—for example, 89% turnout among the traditionalist persona versus a 67% national average, 81% opposition to algorithmic voter targeting, and 94% use of AI image generators. If those numbers are absent or contradicted by the underlying reports, the personas are not data-grounded and the design criteria built on them do not follow.

Watch

Extended reading notes

Core claim

The paper's central claim is that personas produced by an LLM-based deep-research process can serve as the structural backbone of a user-centered visualization workflow for election data. The authors construct three UK personas—the tech-integrated progressive, the privacy-conscious traditionalist, and the digitally disillusioned pragmatist—and show how each persona's technology attitudes, voting behavior, and information-consumption choices map to a dissemination medium, a visual artifact, and the chart types and data to display. They then create three low-fidelity prototypes that embody those design criteria and ask an LLM to summarize and evaluate each against published visualization heuristics and the persona description. In the authors' telling, this demonstrates that LLM-generated personas can make election information easier to understand and more relevant, and that LLM-based evaluation can substitute for direct user participation at the prototyping stage.

Load-bearing premise

The load-bearing premise is that the LLM-generated personas accurately represent real UK voter segments, including the specific statistics in the appended report; if those figures are unverified or invented, the design criteria lose their empirical grounding.

Editorial extensions

If this is right

  • Designers can begin a data-visualization project by generating personas and deriving communication artifacts without running surveys or interviews.
  • The three personas—traditionalist, pragmatist, and progressive—each resolve to a concrete medium: a print infographic, a short data video, and an interactive app.
  • LLM evaluation can be steered by established heuristics, turning the model into a heuristic-evaluation assistant rather than an autonomous critic.
  • The workflow gives visualization researchers a low-cost way to iterate on low-fidelity prototypes before committing to high-fidelity builds.
  • User-centered design claims no longer require direct user participation in every iteration, if personas are accepted as adequate proxies.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • The personas' quoted statistics—such as 89% turnout for one segment and 81% opposition to algorithmic voter targeting—are presented as research findings but trace to an AI deep-research summary, so they should be checked against the underlying survey reports before being reused.
  • The same LLM both proposes design criteria and evaluates the prototypes, meaning the evaluation may be self-consistent rather than independently valid; a human heuristic evaluation of the same prototypes would test whether the AI judge notices what users notice.
  • The mapping from persona attributes to design choices could be tested directly by generating personas with different LLMs or different input data and asking whether the resulting design criteria converge.
  • If the workflow holds, it extends naturally to other civic data—budgets, health statistics, climate indicators—where user segments are similarly diverse but user-research budgets are small.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 8 minor

Summary. The paper proposes a four-stage pipeline for persona-based election visualization: it uses Perplexity deep research to create three UK voter personas (Privacy-Conscious Traditionalist, Digitally Disillusioned Pragmatist, Tech-Integrated Progressive), derives visualization design criteria from persona attributes, builds three prototypes (a print infographic, a TikTok-style data video, and an interactive visualization), and evaluates them by asking ChatGPT to judge them against visualization heuristics and the persona descriptions. The paper concludes with actionable insights for visualization researchers. An appended report, titled 'AI-Driven Voter Personas: Technological Attitudes and Electoral Engagement in the United Kingdom,' provides quantitative statistics attributed to Ipsos, DSIT, and Ofcom and appears to be the raw output of Perplexity deep research.

Significance. If the pipeline were valid, it would address a genuine need: tailoring election visualizations to diverse audience segments without the cost of traditional persona studies. The paper is transparent about its process, provides the persona creation prompt, names all tools used (Perplexity, Datawrapper, Canva, Figma, Adobe Firefly), and describes each prototype concretely enough for others to replicate the pipeline. These are real strengths. However, the central evidence is missing: the persona statistics are not source-verified (and at least one appears factually wrong), the design criteria inherit that unverified foundation, and the LLM-based evaluation is a self-referential narrative rather than a controlled assessment. The work is best read as a speculative design proposal, not as an empirically grounded user-centered method.

major comments (3)
  1. [§4.2 / Appendix after References] The personas' quantitative attributes are presented as research findings, but they are unverified and at least one is demonstrably wrong. The appendix states that Privacy-Conscious Traditionalists have '89% in 2024 vs. 67% national average' turnout, yet the actual 2024 UK general election turnout was approximately 59.9%, not 67%. The paper gives no methodology for how the statistics were extracted from the cited Ipsos, DSIT, and Ofcom documents, no cross-checking protocol, and no statement that the appendix is the raw output of Perplexity deep research. Since §5.1.3 and §5.3.3 explicitly derive design decisions from these statistics (e.g., 'high electoral turnout,' 'local community ties'), the entire persona-to-criteria chain inherits the error. This is load-bearing because Section 4.2 claims the personas have 'empirical validity' while providing no verification of the underlying numbers.
  2. [§7 Prototype Evaluation] The evaluation cannot support the paper's claim that the prototypes make election information 'easier to understand and more relevant.' Section 7 describes asking ChatGPT to summarize each prototype and then judge it against heuristics [54,65] and the persona descriptions. Because the persona descriptions were themselves generated by a similar LLM (Perplexity), this is a self-referential loop: the evaluator is not independent of the design input. There is no baseline, no human participants, no comprehension or recall measure, and no comparison against a non-persona visualization. The quoted outputs in §7.1–7.3 are qualitative narratives (e.g., 'The infographic effectively conveys...'), not measurements. Consequently, the abstract's effectiveness claim is untested. The authors would need either a controlled user study or a clear re-framing of the evaluation as an illustration of the pipeline, not as validation.
  3. [§8.2 Limitations & Future Work / Contribution (3)] The limitation statement 'without direct user participation' directly undercuts the claimed contribution of 'Design, Development & Evaluation' and the paper's repeated use of 'user-centered.' The authors argue that involving selected users is not necessary and that personas carry user perspectives, but this is an assertion, not a demonstrated result. The paper does not show that the personas correspond to real user needs or that the derived criteria improve outcomes. The limitation paragraph itself acknowledges missing support, yet the conclusion (Section 9) still presents the framework as a settled methodology. The authors must either provide external validation or substantially weaken the paper's claims, since the central contribution is currently an untested proposal.
minor comments (8)
  1. [Introduction] The sentence 'the significance of electoral data visualization in molding public opinion is substantial and cannot not be underestimated' contains a double negative; it should read 'cannot be underestimated.'
  2. [§2.1] Reference [4 ??] is malformed; either [4] (FiveThirtyEight) is intended or a reference is missing.
  3. [§4.1] The heading 'T echnology Attitudes' should be 'Technology Attitudes.'
  4. [§2.2] The word 'uncertainity' is misspelled as 'uncertainty.'
  5. [§7] The evaluation is not reproducible: the model version, date, temperature, and full prompt for ChatGPT are not given, and 'temporary ChatGPT mode' is unexplained.
  6. [Figures 3–5] The figure captions are minimal; add a description of each prototype's content so the reader can assess the design choices without guessing from the main text alone.
  7. [Appendix after References] The appended 'AI-Driven Voter Personas' report is not labeled as an appendix or identified as the output of Perplexity deep research; this should be stated explicitly for transparency.
  8. [§8.1 Actionable Insights] In insight (2), 'LLMs enable a intermediate steps' should be 'LLMs enable an intermediate step.'

Circularity Check

1 steps flagged · score 6.0 of 10

LLM-generated personas are both the design input and the evaluation target, so the claimed user-centered validation reduces to an internal LLM consistency check.

  1. self definitional [Section 7 (Prototype Evaluation), final paragraph; combined with Section 4.2 and Fig. 2]
    "To verify the accurate interpretation of the image, the LLM is prompted to generate a descriptive summary of its contents. Subsequently, the LLM is queried to assess whether the visualization aligns with established visualization heuristics [54, 65] and the user persona description (2)."

    The 'user persona description (2)' is the same Perplexity-deep-research output (Section 4.2) from which the Section 5 design considerations and Section 6 prototypes were derived. Thus the evaluation target—'aligned with the persona'—is the exact input that generated the designs. An LLM scoring prototypes against LLM-generated personas is an internal consistency check between two LLM outputs, not evidence that the prototypes meet real UK voter needs. The claimed user-centered validation is defined by the artifact it is supposed to validate; only the separate heuristic checklist [54, 65] provides external content, and it does not ground the persona attributes.

full rationale

The derivation chain is: Perplexity deep research generates three personas (§4.2, Fig. 2), design considerations are derived from those personas (§5), prototypes are built (§6), and LLM evaluation judges the prototypes against 'established visualization heuristics [54, 65] and the user persona description (2)' (§7). The persona description used as the evaluation target is the same LLM-generated text that drove the design, so the evaluation cannot independently establish user-centeredness; it only checks that the prototype matches the LLM's own persona. This is partial circularity. The paper itself concedes that it presents 'a visualization framework that embraces user-centered principles without direct user participation' (§8.2), confirming that no external user ground truth is introduced. External anchors do exist: the real UK election dataset [44], visualization practice, and heuristics [54, 65] provide independent content for the design artifacts. The appended Perplexity report, with statistics such as '89% in 2024 vs. 67% national average' and '81% oppose algorithmic voter targeting', is presented as empirical grounding but is itself an LLM artifact with unverified sources; that is primarily a correctness risk, but it aggravates the self-referential loop because those numbers are the only empirical-looking support for the personas. Self-citation is not a significant issue here: reference [22] overlaps with one author but is a minor comparison-visualization citation, not load-bearing, and no uniqueness theorem is imported. Overall, the central 'evaluate these prototypes using these personas and LLMs' claim reduces by construction to an LLM self-consistency check, but the external heuristic layer and real election data prevent a full 8; the appropriate score is 6.

Assumptions & free parameters 1 free parameters · 6 assumptions · 3 invented entities

The framework rests on a small set of unverified assumptions: that LLM personas correspond to real voter segments, that the appended statistics are true, that generated personas are valid design proxies, and that LLM heuristic evaluation is a credible substitute for human testing. One free parameter is noted: the many persona statistics that act as ad hoc inputs. No independently evidenced entities are introduced; the three personas are fictional design constructs with no external validation.

free parameters (1)
  • Persona demographic/behavioral statistics = Eight-plus percentages, e.g., 89% traditionalist turnout, 81% opposition to algorithmic targeting, 94% pragmatist use…
    The persona report appended after the references supplies these numbers; they are neither measured by the authors nor sourced to a specific table or page. They function as ad hoc inputs that shape the design criteria in Section 5.
assumptions (6)
  • ad hoc to paper The three LLM-generated personas accurately represent real UK voter segments, including their quantified attributes.
    Section 4.2 and the appended persona report treat Perplexity deep research output as empirical data; no independent survey is cited for the specific percentages.
  • domain assumption LLM-generated personas are valid proxies for real users in visualization design.
    The methodology and Section 2.3 argue that LLM personas can replace traditional persona research without user validation; this is an assumption, not a demonstrated equivalence.
  • domain assumption LLM-based heuristic evaluation of visualization images is a valid way to assess design quality.
    Section 7 uses ChatGPT to judge prototypes, but the paper does not compare LLM judgement to human heuristic evaluation or usability testing.
  • domain assumption The UK election dataset from the government website is accurate and properly cited.
    Prototype data is sourced from Ref [44], which is cited as an economic outlook briefing, not an election results dataset; this citation mismatch is unaddressed.
  • domain assumption Established visualization heuristics (Refs 54 and 65) apply directly to election data prototypes.
    The paper applies general infographic and mobile map heuristics to specific election graphics without testing their applicability in this domain.
  • domain assumption The low-fidelity prototypes, rendered as static images, faithfully represent the interactive or video artifacts.
    The data video is shown as four snapshots and the interactive prototype as static images; the LLM evaluation cannot experience animation or interaction.
invented entities (3)
  • Privacy-Conscious Traditionalist persona
    purpose: Represents older, rural, print-media voters so that the paper can derive design criteria for a static infographic.
    The persona is fictional and its statistics come from the LLM report; the paper provides no survey or dataset validating this archetype.
  • Digitally Disillusioned Pragmatist persona
    purpose: Represents younger, social-media-native voters so that the paper can justify a TikTok-style data video.
    No independent evidence is provided that this archetype or its behavioral percentages are accurate.
  • Tech-Integrated Progressive persona
    purpose: Represents urban, tech-savvy voters so that the paper can justify an interactive visualization.
    No independent evidence is provided that this archetype or its behavioral percentages are accurate.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Leveraging LLMs for Persona-Based Visualization of Election Data." pith.science (2026). https://pith.science/paper/O4GUKSPN

@misc{pith2026250721900,
  author       = {Pith},
  title        = {Pith review of: Leveraging LLMs for Persona-Based Visualization of Election Data},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/O4GUKSPN}},
  note         = {Machine review of arXiv:2507.21900}
}
read the original abstract

Visualizations are essential tools for disseminating information regarding elections and their outcomes, potentially influencing public perceptions. Personas, delineating distinctive segments within the populace, furnish a valuable framework for comprehending the nuanced perspectives, requisites, and behaviors of diverse voter demographics. In this work, we propose making visualizations tailored to these personas to make election information easier to understand and more relevant. Using data from UK parliamentary elections and new developments in Large Language Models (LLMs), we create personas that encompass the diverse demographics, technological preferences, voting tendencies, and information consumption patterns observed among voters.Subsequently, we elucidate how these personas can inform the design of visualizations through specific design criteria. We then provide illustrative examples of visualization prototypes based on these criteria and evaluate these prototypes using these personas and LLMs. We finally propose some actionable insights based upon the framework and the different design artifacts.

Figures

Figures reproduced from arXiv: 2507.21900 by the authors.

Figure 1
Figure 1. The Methodology The methodology outlined in this paper employs a systematic approach to designing and evaluating visualization prototypes using User Personas and LLMs. Initially, the process involves leveraging LLMs to generate user personas, which serve as founda￾tional representations of target users. These personas are crafted to encapsulate diverse user characteristics, preferences, and behaviors, providing a ro… view at source ↗
Figure 2
Figure 2. The 3 User Personas. 5. Design Considerations for Visualizations Following the development of user personas, we established specific design considera￾tions for visualization purposes. These considerations include selection of a visualiza￾tion communication artifact, such as an infographic, video presentation, or interactive interface; chart choices and visual design such as maps, bar charts, comparative visu￾alizati… view at source ↗
Figure 3
Figure 3. A infographic prototype. 11 [PITH_FULL_IMAGE:figures/full_fig_p011_3.png] view at source ↗
Figures from the paper (2 more)
Figure 4
Figure 4. Figure 4: Smartphone prototypes [PITH_FULL_IMAGE:figures/full_fig_p012_4.png]
Figure 5
Figure 5. Figure 5: Four snapshots of the Data Video prototype 5.3. For the Tech Integrated Progressive 5.3.1. Visualization Communication Artifact Interactive visualization serves as a powerful tool in academic research and data anal￾ysis, offering significant utility by enhancing user e…

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

72 extracted references · 68 canonical work pages

  1. [1]

    Available at https://www.nytimes.com/ spotlight/graphics

    The New York Times, Graphics (2024). Available at https://www.nytimes.com/ spotlight/graphics

    work page 2024

  2. [2]

    Available at https://www.economist.com/ graphic-detail

    The Economist, Graphic detail (2024). Available at https://www.economist.com/ graphic-detail

    work page 2024

  3. [3]

    I. Alieva, How american media framed 2016 presidential election using data visualization: The case study of the new york times and the washington post , Journalism Practice 17 (2023), pp. 814–840

    work page 2023

  4. [4]

    Silver, Why fivethirtyeight gave trump a better chance than almost anyone else , FiveThirtyEight (2016)

    N. Silver, Why fivethirtyeight gave trump a better chance than almost anyone else , FiveThirtyEight (2016). 17

    work page 2016

  5. [5]

    Towner, The infographic election: The role of visual content on social media in the 2016 presidential campaign , in The presidency and social media , Routledge, 2017, pp

    T.L. Towner, The infographic election: The role of visual content on social media in the 2016 presidential campaign , in The presidency and social media , Routledge, 2017, pp. 236–262

    work page 2016

  6. [6]

    Lurie and C.H

    N.H. Lurie and C.H. Mason, Visual representation: Implications for decision making , Journal of marketing 71 (2007), pp. 160–177

    work page 2007

  7. [7]

    Plaisant, The challenge of information visualization evaluation , in Proceedings of the working conference on Advanced visual interfaces

    C. Plaisant, The challenge of information visualization evaluation , in Proceedings of the working conference on Advanced visual interfaces . 2004, pp. 109–116

    work page 2004

  8. [8]

    L.Y.H. Lo, A. Gupta, K. Shigyo, A. Wu, E. Bertini, and H. Qu, Misinformed by visu- alization: What do we learn from misinformative visualizations? , in Computer Graphics Forum, Vol. 41. Wiley Online Library, 2022, pp. 515–525

    work page 2022

Show all 72 references
  1. [9]

    Correll, Ethical dimensions of visualization research , in Proceedings of the 2019 CHI conference on human factors in computing systems

    M. Correll, Ethical dimensions of visualization research , in Proceedings of the 2019 CHI conference on human factors in computing systems . 2019, pp. 1–13

    2019

  2. [10]

    Mulder and Z

    S. Mulder and Z. Yaar, The user is always right: A practical guide to creating and using personas for the web , New Riders, 2006

    2006

  3. [11]

    Long, Real or imaginary: The effectiveness of using personas in product design , in Proceedings of the Irish Ergonomics Society annual conference, Vol

    F. Long, Real or imaginary: The effectiveness of using personas in product design , in Proceedings of the Irish Ergonomics Society annual conference, Vol. 14. Dublin, 2009, pp. 1–10

    2009

  4. [12]

    Miaskiewicz and K.A

    T. Miaskiewicz and K.A. Kozar, Personas and user-centered design: How can personas benefit product design processes?, Design studies 32 (2011), pp. 417–430

    2011

  5. [13]

    Nielsen, M

    L. Nielsen, M. Larusdottir, and L.B. Larsen, Understanding users through three types of personas, in Human-Computer Interaction–INTERACT 2021: 18th IFIP TC 13 Inter- national Conference, Bari, Italy, August 30–September 3, 2021, Proceedings, Part II 18 . Springer, 2021, pp. 330–348

    2021

  6. [14]

    Pruitt and J

    J. Pruitt and J. Grudin, Personas: practice and theory , in Proceedings of the 2003 con- ference on Designing for user experiences . 2003, pp. 1–15

    2003

  7. [15]

    M. Kay, T. Kola, J.R. Hullman, and S.A. Munson, When (ish) is my bus? user-centered visualizations of uncertainty in everyday, mobile predictive systems , in Proceedings of the 2016 chi conference on human factors in computing systems . 2016, pp. 5092–5103

    2016

  8. [16]

    B¨ ottinger, H.N

    M. B¨ ottinger, H.N. Kostis, M. Velez-Rojas, P. Rheingans, and A. Ynnerman, Reflections on visualization for broad audiences , Foundations of data visualization (2020), pp. 297– 305

    2020

  9. [17]

    De Haan, S

    Y. De Haan, S. Kruikemeier, S. Lecheler, G. Smit, and R. Van der Nat, When does an infographic say more than a thousand words? audience evaluations of news visualizations , Journalism Studies 19 (2018), pp. 1293–1312

    2018

  10. [18]

    Lee, E.K

    B. Lee, E.K. Choe, P. Isenberg, K. Marriott, and J. Stasko, Reaching broader audiences with data visualization , IEEE computer graphics and applications 40 (2020), pp. 82–90

    2020

  11. [19]

    Kim, S.C

    N.W. Kim, S.C. Joyner, A. Riegelhuth, and Y. Kim, Accessible visualization: Design space, opportunities, and challenges , in Computer graphics forum , Vol. 40. Wiley Online Library, 2021, pp. 173–188

    2021

  12. [20]

    Solop and N.A

    F.I. Solop and N.A. Wonders, Data journalism versus traditional journalism in election reporting: an analysis of competing narratives in the 2012 presidential election , Electronic News 10 (2016), pp. 203–223

    2016

  13. [21]

    F. Yang, M. Cai, C. Mortenson, H. Fakhari, A.D. Lokmanoglu, J. Hullman, S. Fran- coneri, N. Diakopoulos, E.C. Nisbet, and M. Kay, Swaying the public? impacts of election forecast visualizations on emotion, trust, and intention in the 2022 us midterms , IEEE Transactions on Vis...

    2023

  14. [22]

    Panda and S.T

    S. Panda and S.T. Roy, Comparing voter electoral preferences using visualizations. , In- ternational Journal of Visual Design 15 (2021)

    2021

  15. [23]

    J. Rao, K. Chen, E.F. Yang, J. Kruse, K. Hudson, and S. Gao, A multi-perspective narrative-based geovisualization dashboard for the 2020 us presidential election , Journal of Geovisualization and Spatial Analysis 5 (2021), pp. 1–15

    2021

  16. [24]

    Yaqub, N

    U. Yaqub, N. Sharma, R. Pabreja, S.A. Chun, V. Atluri, and J. Vaidya, Location-based sentiment analyses and visualization of twitter election data , Digital Government: Re- search and Practice 1 (2020), pp. 1–19. 18

    2020

  17. [25]

    Blomquist and M

    ˚A. Blomquist and M. Arvola, Personas in action: ethnography in an interaction design team, in Proceedings of the second Nordic conference on Human-computer interaction . 2002, pp. 197–200

    2002

  18. [26]

    Pruitt and T

    J. Pruitt and T. Adlin, The persona lifecycle: keeping people in mind throughout product design, Elsevier, 2010

    2010

  19. [27]

    Salminen, K

    J. Salminen, K. Guan, S.g. Jung, S.A. Chowdhury, and B.J. Jansen, A literature review of quantitative persona creation , in Proceedings of the 2020 CHI Conference on Human Factors in Computing Systems . 2020, pp. 1–14

    2020

  20. [28]

    Jansen, S.G

    B.J. Jansen, S.G. Jung, L. Nielsen, K.W. Guan, and J. Salminen, How to create personas: Three persona creation methodologies with implications for practical employment , Pacific Asia Journal of the Association for Information Systems 14 (2022), p. 1

    2022

  21. [29]

    L.C. Koh, A. Slingsby, J. Dykes, and T.S. Kam, Developing and applying a user-centered model for the design and implementation of information visualization tools , in 2011 15th International Conference on Information Visualisation . IEEE, 2011, pp. 90–95

    2011

  22. [30]

    Goodwin, J

    S. Goodwin, J. Dykes, S. Jones, I. Dillingham, G. Dove, A. Duffy, A. Kachkaev, A. Slingsby, and J. Wood, Creative user-centered visualization design for energy analysts and modelers, IEEE transactions on visualization and computer graphics 19 (2013), pp. 2516–2525

    2013

  23. [31]

    Wassink, O

    I. Wassink, O. Kulyk, B.v. Dijk, G.v.d. Veer, and P.v.d. Vet, Applying a user-centered approach to interactive visualisation design , Springer, 2009

    2009

  24. [32]

    Elias and V

    B. Elias and V. Paelke, User-centered design of landmark visualizations , in Map-based mobile services: Design, interaction and usability , Springer, 2008, pp. 33–56

    2008

  25. [33]

    V´ azquez,Are llms ready for visualization? , in 2024 IEEE 17th Pacific Visualization Conference (PacificVis)

    P.P. V´ azquez,Are llms ready for visualization? , in 2024 IEEE 17th Pacific Visualization Conference (PacificVis). IEEE, 2024, pp. 343–352

    2024

  26. [34]

    Lo and H

    L.Y.H. Lo and H. Qu, How good (or bad) are llms at detecting misleading visualizations? , IEEE Transactions on Visualization and Computer Graphics (2024)

    2024

  27. [35]

    J. Hong, C. Seto, A. Fan, and R. Maciejewski, Do llms have visualization literacy? an evaluation on modified visualizations to test generalization in data interpretation , IEEE Transactions on Visualization and Computer Graphics (2025)

    2025

  28. [36]

    Y. Tian, W. Cui, D. Deng, X. Yi, Y. Yang, H. Zhang, and Y. Wu, Chartgpt: Leveraging llms to generate charts from abstract natural language , IEEE Transactions on Visualiza- tion and Computer Graphics (2024)

    2024

  29. [37]

    L. Wang, S. Zhang, Y. Wang, E.P. Lim, and Y. Wang, Llm4vis: Explainable visualization recommendation using chatgpt, arXiv preprint arXiv:2310.07652 (2023)

    2023 arXiv

  30. [38]

    N.W. Kim, G. Myers, and B. Bach, How good is chatgpt in giving advice on your visual- ization design? , arXiv preprint arXiv:2310.09617 (2023)

    2023 arXiv

  31. [39]

    S. De Paoli, Improved prompting and process for writing user personas with llms, using qualitative interviews: Capturing behaviour and personality traits of users , arXiv preprint arXiv:2310.06391 (2023)

    2023 arXiv

  32. [40]

    Huang, T

    Y. Huang, T. Kanij, A. Madugalla, S. Mahajan, C. Arora, and J. Grundy, Unlocking adaptive user experience with generative ai , arXiv preprint arXiv:2404.05442 (2024)

    2024 arXiv

  33. [41]

    Schuller, D

    A. Schuller, D. Janssen, J. Blumenr¨ other, T.M. Probst, M. Schmidt, and C. Kumar, Generating personas using LLMs and assessing their viability , in Extended Abstracts of the CHI Conference on Human Factors in Computing Systems . 2024, pp. 1–7

    2024

  34. [42]

    Schmidt, P

    A. Schmidt, P. Elagroudy, F. Draxler, F. Kreuter, and R. Welsch, Simulating the human in hcd with chatgpt: Redesigning interaction design with ai , Interactions 31 (2024), pp. 24–31

    2024

  35. [43]

    Gerosa, B

    M. Gerosa, B. Trinkenreich, I. Steinmacher, and A. Sarma, Can ai serve as a substitute for human subjects in software engineering research? , Automated Software Engineering 31 (2024), p. 13

    2024

  36. [44]

    Avail- able at https://commonslibrary.parliament.uk/research-briefings/cbp-10009/, accessed: 2025-03-30

    UK Parliament, Research briefing cbp-10009: Uk economic outlook (2023). Avail- able at https://commonslibrary.parliament.uk/research-briefings/cbp-10009/, accessed: 2025-03-30

    2023

  37. [45]

    Siika-aho, et al

    P. Siika-aho, et al. , Persona creation based on secondary data: A study on perceived reli- 19 ability in ux design , Master’s thesis, 2016

    2016

  38. [46]

    Devitt, Creating personas from design ethnography and grounded theory , Journal of Usability Studies 13 (2021)

    F. Devitt, Creating personas from design ethnography and grounded theory , Journal of Usability Studies 13 (2021)

    2021

  39. [47]

    S.G. Jung, J. Salminen, K.K. Aldous, and B.J. Jansen, Personacraft: Leveraging language models for data-driven persona development , International Journal of Human-Computer Studies 197 (2025), p. 103445

    2025

  40. [48]

    Zhang, H.F

    X. Zhang, H.F. Brown, and A. Shankar, Data-driven personas: Constructing archetypal users with clickstreams and user telemetry , in Proceedings of the 2016 CHI conference on human factors in computing systems . 2016, pp. 5350–5359

    2016

  41. [49]

    AI, Perplexity ai: Advanced deep research technology (2025)

    P. AI, Perplexity ai: Advanced deep research technology (2025). Available at https:// www.perplexity.ai, accessed: 2025-03-30

    2025

  42. [50]

    A. Li, H. Chen, H. Namkoong, and T. Peng, Llm generated persona is a promise with a catch, arXiv preprint arXiv:2503.16527 (2025)

    2025 arXiv

  43. [51]

    Shin, M.A

    J. Shin, M.A. Hedderich, B.J. Rey, A. Lucero, and A. Oulasvirta, Understanding human- AI workflows for generating personas , in Proceedings of the 2024 ACM Designing Inter- active Systems Conference . 2024, pp. 757–781

    2024

  44. [52]

    AI, Introducing perplexity deep research (2025)

    P. AI, Introducing perplexity deep research (2025). Available at https://www. perplexity.ai/hub/blog/introducing-perplexity-deep-research , accessed: 2025- 03-30

    2025

  45. [53]

    Available at https://www

    Adobe, Adobe firefly: Generative ai for creativity (2025). Available at https://www. adobe.com/sensei/generative-ai/firefly.html, accessed: 2025-03-30

    2025

  46. [54]

    Siricharoen and N

    W.V. Siricharoen and N. Siricharoen, How infographic should be evaluated, in Proceedings of the 7th International Conference on Information Technology (ICIT 2015) . 2015, pp. 558–564

    2015

  47. [55]

    Crampton, Maps as social constructions: power, communication and visualization , Progress in human Geography 25 (2001), pp

    J.W. Crampton, Maps as social constructions: power, communication and visualization , Progress in human Geography 25 (2001), pp. 235–252

    2001

  48. [56]

    Segel and J

    E. Segel and J. Heer, Narrative visualization: Telling stories with data , IEEE transactions on visualization and computer graphics 16 (2010), pp. 1139–1148

    2010

  49. [57]

    Gleicher, D

    M. Gleicher, D. Albers, R. Walker, I. Jusufi, C.D. Hansen, and J.C. Roberts, Visual comparison for information visualization , Information Visualization 10 (2011), pp. 289– 309

    2011

  50. [58]

    L. Yang, A. Wu, W. Tong, X. Xu, Z. Wei, and H. Qu, Understanding 3d data videos: From screens to virtual reality, in 2023 IEEE 16th Pacific Visualization Symposium (Paci- ficVis). IEEE, 2023, pp. 197–206

    2023

  51. [59]

    B. Lee, M. Brehmer, P. Isenberg, E.K. Choe, R. Langner, and R. Dachselt, Data visual- ization on mobile devices , in Extended Abstracts of the 2018 CHI Conference on Human Factors in Computing Systems . 2018, pp. 1–8

    2018

  52. [60]

    Meyer, A

    J. Meyer, A. Kazakova, M. B¨ using, and S. Boll, Visualization of complex health data on mobile devices , in Proceedings of the 2016 ACM Workshop on Multimedia for Personal Health and Health Care . 2016, pp. 31–34

    2016

  53. [61]

    Coleti, M

    T.A. Coleti, M. Morandini, L. Vilela Leite Filgueiras, P.L. Pizzigatti Correa, I.G. de Oliveira, and C.R.S.C. de Barbosa, Design patterns to support personal data transparency visualization in mobile applications, in Human-Computer Interaction. Perspectives on De- sign: Themat...

    2019

  54. [62]

    Studio, Bar chart race (2025)

    F. Studio, Bar chart race (2025). Available at https://app.flourish.studio/ @flourish/bar-chart-race, accessed: 2025-03-31

    2025

  55. [63]

    D. Keim, G. Andrienko, J.D. Fekete, C. G¨ org, J. Kohlhammer, and G. Melan¸ con,Visual analytics: Definition, process, and challenges , Springer, 2008

    2008

  56. [64]

    Wong and J

    P.C. Wong and J. Thomas, Visual analytics, IEEE Computer Graphics and Applications 24 (2004), pp. 20–21

    2004

  57. [65]

    Kuparinen, J

    L. Kuparinen, J. Silvennoinen, and H. Isom¨ aki, Introducing usability heuristics for mo- bile map applications , in Proceedings of the 26th International Cartographic Conference, 20 August, Vol. 25. 2013, p. 2013

    2013

  58. [66]

    Munzner, Process and pitfalls in writing information visualization research papers , in Information visualization: human-centered issues and perspectives , Springer, 2008, pp

    T. Munzner, Process and pitfalls in writing information visualization research papers , in Information visualization: human-centered issues and perspectives , Springer, 2008, pp. 134–153

    2008

  59. [67]

    Mutlu, E

    B. Mutlu, E. Veas, and C. Trattner, Vizrec: Recommending personalized visualizations , ACM Transactions on Interactive Intelligent Systems (TiiS) 6 (2016), pp. 1–39

    2016

  60. [68]

    Qian, R.A

    X. Qian, R.A. Rossi, F. Du, S. Kim, E. Koh, S. Malik, T.Y. Lee, and N.K. Ahmed, Personalized visualization recommendation, ACM Transactions on the Web (TWEB) 16 (2022), pp. 1–47

    2022

  61. [69]

    Angelucci and A

    C. Angelucci and A. Prat, Is journalistic truth dead? measuring how informed voters are about political news, American Economic Review 114 (2024), pp. 887–925

    2024

  62. [70]

    Linxen, C

    S. Linxen, C. Sturm, F. Br¨ uhlmann, V. Cassau, K. Opwis, and K. Reinecke,How weird is CHI?, in Proceedings of the 2021 chi conference on human factors in computing systems . 2021, pp. 1–14

    2021

  63. [71]

    A. Jena, M. Butler, T. Dwyer, K. Ellis, U. Engelke, R. Kirkham, K. Marriott, C. Paris, and V. Rajamanickam, The next billion users of visualization , IEEE Computer Graphics and Applications 41 (2021), pp. 8–16

    2021

  64. [72]

    info- swaps

    C. Olea, H. Tucker, J. Phelan, C. Pattison, S. Zhang, M. Lieb, D. Schmidt, and J. White, Evaluating persona prompting for question answering tasks . 21 Recent research into voter behavior reveals a complex interplay between technological adoption, information consumption patte...

    2024

Pith tools

Reviewed August 6, 2026 · model on record in the stance chip above.