Pith. sign in

REVIEW 3 major objections 5 minor 75 references

SIA claims coordinated LLM agent flows make heterogeneous social media analysis reliable, efficient, and auditable.

Reviewed by Pith at T0; open to challenge. T0 means a machine referee read the full paper against a public rubric. the ladder, T0–T4 →

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

T0 review

2026-08-04 07:16 UTC pith:YMQ6X7NG

load-bearing objection A credible systems paper with a genuinely new coordinator-plus-taxonomy architecture for heterogeneous social media analysis, but the quantitative evaluation does not back the reliability/efficiency claim. the 3 major comments →

arxiv 2510.26172 v2 pith:YMQ6X7NG submitted 2025-10-30 cs.HC cs.AIcs.SI

Linking Heterogeneous Data with Coordinated Agent Flows for Social Media Analysis

Shifu Chen , Dazhen Deng , Zhihong Xu , Sijia Xu , Linyu Qin , Tai-Quan Peng , Yingcai Wu This is my paper
classification cs.HC cs.AIcs.SI
keywords social media analysisLLM agentsvisual analyticsheterogeneous datainsight taxonomycoordinated agent flowshuman-AI collaborationinformation diffusion
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved

The pith

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

This paper seeks to establish that an LLM-driven multi-agent system, guided by a taxonomy of social-media insight types and a heterogeneity coordinator, can discover meaningful insights from data that mixes user attributes, text, and network structure. Existing automated analysis tools largely handle structured tables and fall short on the messy, multimodal realities of social media. The paper argues that a taxonomy linking what analysts want to know with appropriate mining and visualization methods, plus a coordinator that keeps tabular, textual, and network data flowing coherently through the pipeline, turns a general analysis goal into a traceable, inspectable workflow. If right, this means non-experts could run complex social media analyses and trust the results enough to question them.

Core claim

The paper's central claim is that SIA, by pairing a taxonomy of social-media insight types with a heterogeneity coordinator that keeps tabular, textual, and network data flowing coherently through a multi-agent pipeline, can discover diverse and meaningful insights across these modalities while preserving an auditable trail from goal to report. The system decomposes a goal into query, mining, visualization, and reporting stages; the planner maintains a full path history so each step is informed by prior context, and the coordinator adapts data formats and links entities through shared identifiers. The paper supports this claim with two expert-centered case studies, one on 2020 U.S. election

What carries the argument

The load-bearing mechanism is a two-part design: (1) a bottom-up taxonomy of social-media insight types — organized by entity (single user, user group, single UGC, UGC group) and by static versus dynamic temporal character — that maps each insight type to representative mining methods and visualization strategies; and (2) a heterogeneity coordinator that unifies tabular data, text, and network structure through shared identifiers, transforming outputs into the input formats each downstream agent needs. The planner, which keeps a path history of actions, results, interpretations, and next-step suggestions, uses the taxonomy to choose directions and agents, while the coordinator ensures the da

Load-bearing premise

The load-bearing premise is that the proposed taxonomy is a complete and correct mapping of the insight types analysts actually seek to the mining and visualization methods that address them; if it misses common insight types or points to the wrong method, the planner will systematically choose poor analyses even when every agent executes correctly.

What would settle it

Hold out a set of published social-media case studies that were not used to build the taxonomy, ask domain experts to state the intended insight type and the ideal mining and visualization method for each case, then compare those with SIA's planner selections; if the planner misses routine insight types or systematically disagrees with experts on method choice, the taxonomy's completeness and correctness claims fail.

Watch this falsifier. Get emailed when new claim-graph text bears on it.

If this is right

  • A user can pose a natural-language goal and receive a structured report whose insights link back to inspectable agent-tree nodes, so findings can be checked and refined rather than taken on faith.
  • The data coordinator connects tabular attributes, text, and network structure via shared identifiers, enabling analyses that combine modalities without manual data wrangling.
  • The planner regularly proposes multiple exploration directions for one goal; both experts in the case studies reported that the proposed directions covered approaches they had not initially considered.
  • Across several language models, action-level error rates stay below 12 percent and most failed actions recover within one or two retries, supporting the reliability claim.
  • A dedicated mining view exposes parameter–quality relationships in parallel coordinates, letting users judge whether a mined pattern is stable before accepting it.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • Editorial inference — the taxonomy's completeness is the main untested load: it was built from a selected corpus and refined with only two experts, so a broader set of domain experts could test whether the categories actually cover the questions analysts ask.
  • Editorial inference — the path-based context mechanism deliberately isolates parallel exploration branches, so insights that would require combining evidence across two branches are not available to the planner; a cross-path memory mechanism is a natural extension.
  • Editorial inference — the same coordinator-plus-taxonomy pattern transfers to other heterogeneous analysis domains, such as health records combining structured vitals, clinical text, and social graphs, with the taxonomy replaced by a domain-specific insight map.
  • Editorial inference — a quantitative way to test the taxonomy's robustness is to measure inter-rater agreement when new experts assign open-ended insight descriptions to the Table 1 categories; low agreement would indicate the categories are not stable.

Editorial analysis

A structured set of objections, weighed in public.

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

Referee Report

3 major / 5 minor

Summary. The paper presents SIA (Social Insight Agents), an LLM-agent system for exploratory social media analysis. SIA combines an insight taxonomy, a planner that decomposes user goals, query/mining/visualization/report agents, and a heterogeneity coordinator that links tabular, textual, and network data through shared identifiers. The system is implemented on TwiBot-22 and includes an interactive interface. The authors claim that, guided by the taxonomy and coordinated agent flows, SIA can discover diverse and meaningful insights from heterogeneous social media data while improving reliability and efficiency, and they support this with two expert case studies and a quantitative evaluation of LLM action latency and error rate.

Significance. If the central claims were established, SIA would be a meaningful step beyond existing LLM-based insight-discovery systems that are largely confined to structured tabular data. The strengths of the paper are the clear formalization of the agent workflow (Eqs. 1-16), the bottom-up construction of a social-media insight taxonomy, the design of the heterogeneity coordinator, and an interface with explicit traceability and steering mechanisms. The framework is coherent and the qualitative case-study material is suggestive. However, the evaluation as presented is not commensurate with the load-bearing claims in the abstract and conclusion: it measures LLM response time and action error rate, not the correctness, novelty, or end-to-end efficiency of discovered insights. The stress-test concern that the central claim is not directly evaluated is well-founded.

major comments (3)
  1. [Section 10; Abstract; Section 12] The quantitative evaluation measures only LLM action-level response time and error rate for 'plan' and 'invoke' actions, with no baseline, no ground truth, and no end-to-end system metric. The 5 tasks x 3 runs describe the underlying LLM API behavior, not SIA's integrated contribution. Computational execution time is explicitly excluded, yet the abstract claims SIA 'enhances both the reliability and efficiency' of social media analysis. To support this claim, the paper needs a comparison against at least one reasonable alternative (e.g., a non-agent pipeline, existing LLM systems such as InsightPilot or LightVA, or a human-analyst baseline), plus end-to-end wall-clock time and an assessment of output-level insight quality. The uncertainty/evaluation formulas (Eqs. 11-16) are presented as quantifying reliability, but no validation or sensitivity analysis of the hand-specified lambda weigh
  2. [Section 4.1; Table 1] The insight taxonomy is load-bearing: the planner selects mining methods and visualization strategies based on Table 1. However, the validation in Section 4.1 consists of refinement by two domain experts; no systematic coverage or completeness check is reported, no inter-rater reliability, and no comparison with alternative taxonomies. If common insight types are missing or method mappings are misaligned, the planner will systematically choose inappropriate analyses even when every agent executes flawlessly. The paper needs a stronger validation of the taxonomy, for example an independent coding study on a held-out corpus, or an ablation showing that removing/replacing the taxonomy degrades output quality.
  3. [Section 9] The two expert case studies provide qualitative endorsement and useful design feedback, but they do not systematically measure the diversity or meaningfulness of the discovered insights. Each session lasted about one hour, the outputs were pre-run, and the reported evidence consists largely of favorable quotes from the two experts. There is no structured scoring of insight correctness, relevance, or novelty, no protocol for negative cases, and no analysis of disagreement between experts. Given that the abstract's central claim is that SIA discovers 'diverse and meaningful insights,' this evidence is suggestive but not sufficient. A more rigorous protocol—such as independent expert ratings of final reports against a rubric, or comparison with expert-generated analyses—is needed.
minor comments (5)
  1. [Section 9.1] Typo: 'Anther expert' should be 'Another expert.' Elsewhere in Section 9.4 there are grammatical issues ('experts indicated high system usability, they noted...').
  2. [Eqs. 11-16] The lambda symbols are reused with different meanings in different subsections. Please define each lambda, state its range and chosen value, and clarify how the weights were selected.
  3. [Figure 6] The figure reports error bars, but the number of trials and the unit of aggregation (actions vs. runs vs. tasks) are not stated in the caption or text. Please clarify the sample sizes underlying the means and standard deviations.
  4. [Section 8] The implementation section gives no reproducibility artifact (code, model versions, or dataset access details beyond the TwiBot-22 citation). Providing a link to code or at least specifying exact model versions and API dates would improve reproducibility.
  5. [Table 1] Some cells are marked 'N/A' (e.g., Static/Dynamic for single UGC content features). A brief explanation of why these entries are not applicable, and how the planner handles them, would help readers understand the taxonomy's coverage.

Circularity Check

0 steps flagged

No circular derivation; the taxonomy is an inductive design artifact and the evaluation rests on external experts and datasets.

full rationale

No circular step is present. SIA's central contribution is a taxonomy and an agent framework, not a fitted prediction: the λ weights in Eqs. 11–16 are hand-specified heuristics, and the paper does not fit them to the evaluation data and then re-predict that data. The taxonomy in Table 1 is constructed bottom-up from prior published case studies and refined with two outside experts (Section 4.1); it is a codification of existing practice, not a result derived from SIA's own outputs. The quantitative evaluation measures LLM action-level error rates and response times with no baseline, and the case studies are qualitative expert sessions; these support or fail to support the effectiveness claim, but they do not reduce that claim to the system's design by construction. Some taxonomy examples cite the authors' own prior systems (e.g., D-Map [11], R-Map [13], EvoRiver [56], OpinionFlow [62]), but those citations are descriptive illustrations backed by published peer-reviewed work and are not load-bearing uniqueness claims. Section 11 explicitly concedes that broader validation is still needed, which is an evidence limitation rather than a circularity. The claimed 'reliability and efficiency' enhancement is under-supported by the evaluation setup, but that is a correctness/evidence gap, not a circular derivation.

Axiom & Free-Parameter Ledger

4 free parameters · 5 axioms · 0 invented entities

The central claim rests on hand-weighted LLM self-evaluation functions (Eqs. 11, 12, 14, 16), a taxonomy built from a selected corpus and two experts, the assumption that shared IDs make heterogeneous stores linkable, and the proxy that LLM error rate/latency reflects analytical effectiveness. No code or external benchmark independently confirms these choices.

free parameters (4)
  • λ1, λ2, λ3 (mining evaluation weights, Eq. 11)
    Weights combine stability, domain metric, and LLM assessment; values not reported and appear hand-chosen.
  • λ4, λ5 (uncertainty weights, Eq. 12)
    Weights for method suitability and evaluation reliability are unspecified; no estimation procedure is given.
  • λ1, λ2, λ3 (visualization evaluation weights, Eq. 14)
    Weights for quality, alignment, and accumulated uncertainty are unspecified and likely tuned by hand.
  • λ1, λ2 (report evaluation weights, Eq. 16)
    Weights for relevance and completeness in report quality are unspecified.
axioms (5)
  • domain assumption Shared user IDs and post IDs are consistent enough across tabular, text, and network stores to enable cross-modal linking.
    Section 3 states inputs are inherently linked through shared identifiers; if IDs are inconsistent, the coordinator's integration fails.
  • domain assumption The bottom-up taxonomy built from selected survey papers and two experts is a sufficient guide for planning social media analyses.
    Section 4.1 describes construction and expert validation; the planner's method selection relies entirely on this taxonomy.
  • domain assumption LLM-based self-evaluation scores are valid proxies for analysis quality.
    Eqs. 11-16 use LLM scores to select mining parameters, visualizations, and reports; no external ground truth calibrates these scores.
  • domain assumption LLM error rate and response time are meaningful proxies for the system's analytical effectiveness.
    Section 10 uses action-level error rate and latency as the primary quantitative outcome, but does not measure insight correctness or user value.
  • domain assumption The 5W framework is an appropriate schema for presenting social media insights.
    Section 7.4 and case-study feedback endorse 5W, but only two experts contributed opinions.

pith-pipeline@v1.3.0-alltime-deepseek · 18750 in / 12830 out tokens · 127381 ms · 2026-08-04T07:16:46.217604+00:00 · methodology

0 comments
read the original abstract

Social media platforms generate volumes of heterogeneous data, capturing user behaviors, textual content, and network structures. Analyzing such data is crucial for understanding phenomena such as opinion dynamics, community formation, and information diffusion. However, discovering insights from this complex landscape is exploratory, conceptually challenging, and requires expertise in social media mining and visualization. Existing automated approaches, including large language models (LLMs), remain largely confined to structured tabular data and cannot adequately address the heterogeneity of social media analysis. We present SIA (Social Insight Agents), an LLM agent system that links heterogeneous multi-modal data, including raw inputs (e.g., text, network, and behavioral data), mined analytical results, and rendered visual artifacts, through coordinated agent flows. Guided by an insight-oriented taxonomy connecting insight types with suitable mining methods and visualization strategies, SIA adopts a stage-synchronized strategy that proceeds through goal decomposition, query, mining, visualization, and reporting stages. At each stage, it collects prior information to jointly plan and execute agent actions, while the coordinator maintains cross-stage action dependencies and assembles and distributes data to agents. Through quantitative evaluation and case studies supported by an interactive interface, we show that SIA can discover diverse and meaningful insights from social media with opportunities for subsequent reliability assessment.

Figures

Figures reproduced from arXiv: 2510.26172 by Dazhen Deng, Linyu Qin, Shifu Chen, Sijia Xu, Tai-Quan Peng, Yingcai Wu, Zhihong Xu.

Figure 1
Figure 1. Figure 1: Overview of SIA. The planner decomposes user goals into ac [PITH_FULL_IMAGE:figures/full_fig_p005_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Overview of the planner. It decomposes a user goal into exploration directions, invokes query, mining, and visualization agents with API [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Role of the heterogeneity coordinator. This component manages data heterogeneity across agents by transforming outputs into required [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: System interface. Chat Panel (C) facilitates dialogue between users and the agent. Action View (B) selectively displays the agent’s actions [PITH_FULL_IMAGE:figures/full_fig_p008_4.png] view at source ↗
Figure 5
Figure 5. Figure 5: Temporal Analysis of COVID-19 social media discussions. The [PITH_FULL_IMAGE:figures/full_fig_p008_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Model performance comparison showing response times and [PITH_FULL_IMAGE:figures/full_fig_p009_6.png] view at source ↗

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Reference graph

Works this paper leans on

75 extracted references · 9 canonical work pages · 1 internal anchor

  1. [1]

    M. S. Aljabri, R. Zagrouba, A. Shaahid, F. Alnasser, A. Saleh, and D. M. Alomari. Machine learning-based social media bot detection: a compre- hensive literature review.Social Network Analysis and Mining, 13(1):20,

  2. [2]

    R. Amar, J. Eagan, and J. Stasko. Low-level components of analytic activity in information visualization. InIEEE Symposium on Information Visualization, 2005. INFOVIS 2005., pp. 111–117, 2005. doi: 10.1109/ INFVIS.2005.1532136 3

    arXiv 2005

  3. [3]

    Aminikhanghahi and D

    S. Aminikhanghahi and D. J. Cook. A survey of methods for time series change point detection.Knowledge and information systems, 51(2):339– 367, 2017. 3

    2017

  4. [4]

    Bakshy, J

    E. Bakshy, J. M. Hofman, W. A. Mason, and D. J. Watts. Everyone’s an influencer: quantifying influence on twitter. In I. King, W. Nejdl, and H. Li, eds.,Proceedings of the Forth International Conference on Web Search and Web Data Mining, WSDM 2011, Hong Kong, China, February 9-12, 2011, pp. 65–74. ACM, 2011. doi: 10.1145/1935826.1935845 4

    arXiv 2011

  5. [5]

    D. M. Blei, A. Y . Ng, and M. I. Jordan. Latent dirichlet allocation.Journal of machine Learning research, 3(Jan):993–1022, 2003. 4

    2003

  6. [6]

    Brandes and B

    U. Brandes and B. Nick. Asymmetric relations in longitudinal social networks.IEEE Transactions on Visualization and Computer Graphics, 17(12):2283–2290, 2011. doi: 10.1109/TVCG.2011.169 4

    work page doi:10.1109/tvcg.2011.169 2011

  7. [7]

    Brehmer and T

    M. Brehmer and T. Munzner. A multi-level typology of abstract visualiza- tion tasks.IEEE Transactions on Visualization and Computer Graphics, 19(12):2376–2385, 2013. doi: 10.1109/TVCG.2013.124 3

    doi:10.1109/tvcg.2013.124 2013

  8. [8]

    N. Cao, Y . Lin, X. Sun, D. Lazer, S. Liu, and H. Qu. Whisper: Tracing the spatiotemporal process of information diffusion in real time.IEEE Transactions on Visualization and Computer Graphics, 18(12):2649–2658,

  9. [9]

    M. Cha, H. Haddadi, F. Benevenuto, and P. K. Gummadi. Measuring user influence in twitter: The million follower fallacy. In W. W. Cohen and S. Gosling, eds.,Proceedings of the Fourth International Conference on Weblogs and Social Media, ICWSM 2010, Washington, DC, USA, May 23-26, 2010. The AAAI Press, 2010. 4

    2010

  10. [10]

    Chauhan and A

    U. Chauhan and A. Shah. Topic modeling using latent dirichlet allocation: A survey.ACM Computing Surveys, 54(7):145:1–145:35, 2022. doi: 10. 1145/3462478 2

    2022

  11. [12]

    S. Chen, S. Feng, S. Liang, C. Zong, J. Li, and P. Li. CACL: community- aware heterogeneous graph contrastive learning for social media bot de- tection. In L. Ku, A. Martins, and V . Srikumar, eds.,Findings of the Association for Computational Linguistics, ACL 2024, Bangkok, Thailand and virtual meeting, August 11-16, 2024, pp. 10349–10360. Association fo...

    doi:10.18653/v1/2024.findings 2024

  12. [13]

    S. Chen, S. Li, S. Chen, and X. Yuan. R-map: A map metaphor for visual- izing information reposting process in social media.IEEE Transactions on Visualization and Computer Graphics, 26(1):1204–1214, 2020. doi: 10 .1109/TVCG.2019.2934263 1, 2, 4

    arXiv 2020

  13. [14]

    S. Chen, L. Lin, and X. Yuan. Social media visual analytics.Computer Graphics Forum, 36(3):563–587, 2017. doi: 10.1111/CGF.13211 1, 2, 3

    work page doi:10.1111/cgf.13211 2017

  14. [15]

    Cheng, L

    J. Cheng, L. A. Adamic, P. A. Dow, J. M. Kleinberg, and J. Leskovec. Can cascades be predicted? In C. Chung, A. Z. Broder, K. Shim, and T. Suel, eds.,23rd International World Wide Web Conference, WWW ’14, Seoul, Republic of Korea, April 7-11, 2014, pp. 925–936. ACM, 2014. doi: 10. 1145/2566486.2567997 4

    arXiv 2014

  15. [16]

    Davidson, D

    T. Davidson, D. Warmsley, M. W. Macy, and I. Weber. Automated hate speech detection and the problem of offensive language. InProceedings of the Eleventh International Conference on Web and Social Media, ICWSM 2017, Montréal, Québec, Canada, May 15-18, 2017, pp. 512–515. AAAI Press, 2017. 4

    2017

  16. [17]

    Devlin, M.-W

    J. Devlin, M.-W. Chang, K. Lee, and K. Toutanova. Bert: Pre-training of deep bidirectional transformers for language understanding. InPro- ceedings of the 2019 conference of the North American chapter of the association for computational linguistics: human language technologies, volume 1 (long and short papers), pp. 4171–4186, 2019. 3

    2019

  17. [18]

    V . Dibia. LIDA: A tool for automatic generation of grammar-agnostic visu- alizations and infographics using large language models. In D. Bollegala, R. Huang, and A. Ritter, eds.,Proceedings of the 61st Annual Meet- ing of the Association for Computational Linguistics (Volume 3: System Demonstrations), pp. 113–126. Association for Computational Linguistic...

    doi:10.18653/v1/2023.acl-demo.11 2023

  18. [19]

    K. Ding, R. Wang, and S. Wang. Social media popularity prediction: A multiple feature fusion approach with deep neural networks. In L. Amsaleg, B. Huet, M. A. Larson, G. Gravier, H. Hung, C. Ngo, and W. T. Ooi, eds., Proceedings of the 27th ACM International Conference on Multimedia, MM 2019, Nice, France, October 21-25, 2019, pp. 2682–2686. ACM,

    2019

  19. [20]

    R. Ding, S. Han, Y . Xu, H. Zhang, and D. Zhang. Quickinsights: Quick and automatic discovery of insights from multi-dimensional data. In P. A. Boncz, S. Manegold, A. Ailamaki, A. Deshpande, and T. Kraska, eds., Proceedings of the 2019 International Conference on Management of Data, SIGMOD Conference 2019, Amsterdam, The Netherlands, June 30 - July 5, 201...

    arXiv 2019

  20. [21]

    D. Edge, N. H. Riche, J. Larson, and C. White. Beyond tasks: An activity typology for visual analytics.IEEE Transactions on Visualization and Computer Graphics, 24(1):267–277, 2018. doi: 10.1109/TVCG.2017. 2745180 3

    doi:10.1109/tvcg.2017 2018

  21. [22]

    S. Feng, Z. Tan, H. Wan, N. Wang, Z. Chen, B. Zhang, Q. Zheng, W. Zhang, Z. Lei, S. Yang, X. Feng, Q. Zhang, H. Wang, Y . Liu, Y . Bai, H. Wang, Z. Cai, Y . Wang, L. Zheng, Z. Ma, J. Li, and M. Luo. Twibot-22: Towards graph-based twitter bot detection. In S. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, and A. Oh, eds.,Advances in Neural Informatio...

    2022

  22. [23]

    E. R. Gansner, Y . Hu, and S. C. North. Interactive visualization of stream- ing text data with dynamic maps.Journal of Graph Algorithms and Applications, 17(4):515–540, 2013. doi: 10.7155/JGAA.00302 4

    work page doi:10.7155/jgaa.00302 2013

  23. [24]

    S. A. Golder and M. W. Macy. Diurnal and seasonal mood vary with work, sleep, and daylength across diverse cultures.Science, 333(6051):1878– 1881, 2011. 4

    2011

  24. [25]

    Y . Gong, Q. Zhang, and X. Huang. Hashtag recommendation for multi- modal microblog posts.Neurocomputing, 272:170–177, 2018. 3

    2018

  25. [26]

    Havre, E

    S. Havre, E. G. Hetzler, and L. T. Nowell. Themeriver: Visualizing theme changes over time. In J. D. Mackinlay, S. F. Roth, and D. A. Keim, eds., IEEE Symposium on Information Visualization 2000 (INFOVIS’00), Salt Lake City, Utah, USA, October 9-10, 2000, pp. 115–123. IEEE Computer Society, 2000. doi: 10.1109/INFVIS.2000.885098 4

    arXiv 2000

  26. [27]

    Heer and danah boyd

    J. Heer and danah boyd. Vizster: Visualizing online social networks. In J. T. Stasko and M. O. Ward, eds.,IEEE Symposium on Information Visualization (InfoVis 2005), 23-25 October 2005, Minneapolis, MN, USA, pp. 32–39. IEEE Computer Society, 2005. doi: 10.1109/INFVIS.2005. 1532126 2

    doi:10.1109/infvis.2005 2005

  27. [28]

    M. Hu, K. Wongsuphasawat, and J. T. Stasko. Visualizing social media content with sententree.IEEE Transactions on Visualization and Computer Graphics, 23(1):621–630, 2017. doi: 10.1109/TVCG.2016.2598590 2, 4

    arXiv 2017

  28. [29]

    Injadat, F

    M. Injadat, F. Salo, and A. B. Nassif. Data mining techniques in social media: A survey.Neurocomputing, 214:654–670, 2016. doi: 10.1016/J. NEUCOM.2016.06.045 1, 3

    doi:10.1016/j 2016

  29. [30]

    Kong, G.-Y

    Y .-X. Kong, G.-Y . Shi, R.-J. Wu, and Y .-C. Zhang. k-core: Theories and applications.Physics Reports, 832:1–32, 2019. 3

    2019

  30. [31]

    Küçük and F

    D. Küçük and F. Can. Stance detection: A survey.ACM Computing Surveys (CSUR), 53(1):1–37, 2020. 1, 3, 4

    2020

  31. [32]

    Kulkarni, R

    V . Kulkarni, R. Al-Rfou, B. Perozzi, and S. Skiena. Statistically significant detection of linguistic change. InProceedings of the 24th international conference on world wide web, pp. 625–635, 2015. 3

    2015

  32. [33]

    H. Kwak, C. Lee, H. Park, and S. B. Moon. What is twitter, a social net- work or a news media? In M. Rappa, P. Jones, J. Freire, and S. Chakrabarti, eds.,Proceedings of the 19th International Conference on World Wide Web, WWW 2010, Raleigh, North Carolina, USA, April 26-30, 2010, pp. 591–600. ACM, 2010. doi: 10.1145/1772690.1772751 4

    arXiv 2010

  33. [34]

    Lee and K

    R. Lee and K. Sumiya. Measuring geographical regularities of crowd behaviors for twitter-based geo-social event detection. In X. Zhou, W. Lee, W. Peng, and X. Xie, eds.,Proceedings of the 2010 International Workshop on Location Based Social Networks, LBSN 2010, November 2, 2010, San Jose, CA, USA, Proceedings, pp. 1–10. ACM, 2010. doi: 10.1145/1867699 .1867701 4

    work page doi:10.1145/1867699 2010

  34. [35]

    Y . Li, J. Fan, Y . Wang, and K. Tan. Influence maximization on social graphs: A survey.IEEE Transactions on Knowledge and Data Engi- neering, 30(10):1852–1872, 2018. doi: 10.1109/TKDE.2018.2807843 2

    arXiv 2018

  35. [36]

    X. Lin, P. R. Spence, and K. A. Lachlan. Social media and credibility indicators: The effect of influence cues.Computers in Human Behavior, 63:264–271, 2016. doi: 10.1016/J.CHB.2016.05.002 4

    work page doi:10.1016/j.chb.2016.05.002 2016

  36. [37]

    Y . Lin, H. Li, L. Yang, A. Wu, and H. Qu. Inksight: Leveraging sketch interaction for documenting chart findings in computational notebooks. IEEE Transactions on Visualization and Computer Graphics, 30(1):944– 954, 2024. doi: 10.1109/TVCG.2023.3327170 1

    arXiv 2024

  37. [38]

    S. Liu, S. Wang, T. Chang, W. Lin, C. Hsiung, Y . Hsieh, Y . Cheng, S. Luo, and J. Zhang. Jarvix: A LLM no code platform for tabular data analysis and optimization. In M. Wang and I. Zitouni, eds.,Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: EMNLP 2023 - Industry Track, Singapore, December 6-10, 2023, pp. 622–

    2023

  38. [39]

    X. Liu, A. Xu, L. Gou, H. Liu, R. Akkiraju, and H. Shen. Socialbrands: Visual analysis of public perceptions of brands on social media. In G. L. Andrienko, S. Liu, and J. T. Stasko, eds.,11th IEEE Conference on Visual Analytics Science and Technology, IEEE VAST 2016, Baltimore, MD, USA, October 23-28, 2016, pp. 71–80. IEEE Computer Society, 2016. doi: 10....

    arXiv 2016

  39. [40]

    P. Ma, R. Ding, S. Han, and D. Zhang. Metainsight: Automatic discovery of structured knowledge for exploratory data analysis. In G. Li, Z. Li, S. Idreos, and D. Srivastava, eds.,SIGMOD ’21: International Conference on Management of Data, Virtual Event, China, June 20-25, 2021, pp. 1262–1274. ACM, 2021. doi: 10.1145/3448016.3457267 2

    arXiv 2021

  40. [41]

    P. Ma, R. Ding, S. Wang, S. Han, and D. Zhang. Insightpilot: An llm- empowered automated data exploration system. In Y . Feng and E. Lefever, eds.,Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, EMNLP 2023 - System Demonstrations, Singapore, December 6-10, 2023, pp. 346–352. Association for Computational Lin- guisti...

    doi:10.18653/v1/2023.emnlp-demo.31 2023

  41. [42]

    Maddigan and T

    P. Maddigan and T. Susnjak. Chat2vis: Generating data visualisations via natural language using chatgpt, codex and GPT-3 large language models. CoRR, abs/2302.02094, 2023. doi: 10.48550/ARXIV.2302.02094 2

    Pith/arXiv arXiv doi:10.48550/arxiv.2302.02094 2023

  42. [43]

    Magelinski, L

    T. Magelinski, L. H. X. Ng, and K. M. Carley. A synchronized action framework for responsible detection of coordination on social media.arXiv preprint arXiv:2105.07454, 2021. 3

    Pith/arXiv arXiv 2021

  43. [44]

    Marjan, N

    M. Marjan, N. Zaki, and E. A. Mohamed. Link prediction in dynamic social networks: A literature review. In2018 IEEE 5th international congress on information science and technology (CIST), pp. 200–207. IEEE, 2018. 3

    2018

  44. [45]

    M. M. Muñoz, M.-M. Rojas-de Gracia, and C. Navas-Sarasola. Measuring engagement on twitter using a composite index: An application to social media influencers.Journal of Informetrics, 16(4):101323, 2022. 3

    2022

  45. [46]

    N. P. Nguyen, T. N. Dinh, Y . Shen, and M. T. Thai. Dynamic social community detection and its applications.PloS one, 9(4):e91431, 2014. 3

    2014

  46. [47]

    B. Pang, L. Lee, et al. Opinion mining and sentiment analysis.Foundations and Trends® in information retrieval, 2(1–2):1–135, 2008. 2, 4

    2008

  47. [48]

    Pennacchiotti and A

    M. Pennacchiotti and A. Popescu. A machine learning approach to twitter user classification. In L. A. Adamic, R. Baeza-Yates, and S. Counts, eds., Proceedings of the Fifth International Conference on Weblogs and Social Media, Barcelona, Catalonia, Spain, July 17-21, 2011. The AAAI Press,

    2011

  48. [49]

    D. Ren, X. Zhang, Z. Wang, J. Li, and X. Yuan. Weiboevents: A crowd sourcing weibo visual analytic system. In I. Fujishiro, U. Brandes, H. Ha- gen, and S. Takahashi, eds.,IEEE Pacific Visualization Symposium, Paci- ficVis 2014, Yokohama, Japan, March 4-7, 2014, pp. 330–334. IEEE Computer Society, 2014. doi: 10.1109/PACIFICVIS.2014.38 4

    work page doi:10.1109/pacificvis.2014.38 2014

  49. [50]

    Schulz, T

    H.-J. Schulz, T. Nocke, M. Heitzler, and H. Schumann. A design space of visualization tasks.IEEE Transactions on Visualization and Computer Graphics, 19(12):2366–2375, 2013. doi: 10.1109/TVCG.2013.120 3

    doi:10.1109/tvcg.2013.120 2013

  50. [51]

    Senevirathna, C

    C. Senevirathna, C. Gunaratne, W. Rand, C. Jayalath, and I. Garibay. Influence cascades: Entropy-based characterization of behavioral influence patterns in social media.Entropy, 23(2):160, 2021. 3

    2021

  51. [52]

    M. A. Shareef, K. K. Kapoor, B. Mukerji, R. Dwivedi, and Y . K. Dwivedi. Group behavior in social media: Antecedents of initial trust formation. Computers in Human Behavior, 105:106225, 2020. 3

    2020

  52. [53]

    Sharma, Y

    K. Sharma, Y . Zhang, E. Ferrara, and Y . Liu. Identifying coordinated accounts on social media through hidden influence and group behaviours. In F. Zhu, B. C. Ooi, and C. Miao, eds.,KDD ’21: The 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Virtual Event, Singapore, August 14-18, 2021, pp. 1441–1451. ACM, 2021. doi: 10. 1145/3447548....

    arXiv 2021

  53. [54]

    Sidorov, A

    S. Sidorov, A. Faizliev, and S. Tikhonova. An extension of the susceptible– infected model and its application to the analysis of information dissemi- nation in social networks.Modelling, 4(4):585–599, 2023. 3

    2023

  54. [55]

    X. Su, S. Xue, F. Liu, J. Wu, J. Yang, C. Zhou, W. Hu, C. Paris, S. Nepal, D. Jin, Q. Z. Sheng, and P. S. Yu. A comprehensive survey on community detection with deep learning.IEEE Transactions on Neural Networks and Learning Systems, 35(4):4682–4702, 2024. doi: 10.1109/TNNLS.2021. 3137396 2, 3

    doi:10.1109/tnnls.2021 2024

  55. [57]

    C. R. Sunstein.Republic: Divided democracy in the age of social media. Princeton University Press, Princeton, NJ, 2018. 4

    2018

  56. [58]

    Trunfio and S

    M. Trunfio and S. Rossi. Conceptualising and measuring social media engagement: A systematic literature review.Italian Journal of Marketing, 2021(3):267–292, 2021. 4

    2021

  57. [59]

    Ugander, B

    J. Ugander, B. Karrer, L. Backstrom, and C. Marlow. The anatomy of the facebook social graph.CoRR, abs/1111.4503, 2011. 4

    Pith/arXiv arXiv 2011

  58. [60]

    L. Wang, S. Zhang, Y . Wang, E. Lim, and Y . Wang. Llm4vis: Explainable visualization recommendation using chatgpt. In M. Wang and I. Zitouni, eds.,Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: EMNLP 2023 - Industry Track, Singapore, Decem- ber 6-10, 2023, pp. 675–692. Association for Computational Linguistics,

    2023

  59. [61]

    Y . Wu, N. Cao, D. Gotz, Y . Tan, and D. A. Keim. A survey on vi- sual analytics of social media data.IEEE Transactions on Multimedia, 18(11):2135–2148, 2016. doi: 10.1109/TMM.2016.2614220 3

    arXiv 2016

  60. [62]

    Y . Wu, S. Liu, K. Yan, M. Liu, and F. Wu. Opinionflow: Visual analysis of opinion diffusion on social media.IEEE Transactions on Visualization and Computer Graphics, 20(12):1763–1772, 2014. doi: 10.1109/TVCG. 2014.2346920 1, 2, 4

    arXiv 2014

  61. [63]

    P. Xu, Y . Wu, E. Wei, T. Peng, S. Liu, J. J. H. Zhu, and H. Qu. Visual analysis of topic competition on social media.IEEE Transactions on Visualization and Computer Graphics, 19(12):2012–2021, 2013. doi: 10. 1109/TVCG.2013.221 1, 2, 4

    2012

  62. [64]

    Yang, M.-H

    J.-A. Yang, M.-H. Tsou, C.-T. Jung, C. Allen, B. H. Spitzberg, J. M. Gawron, and S.-Y . Han. Social media analytics and research testbed (smart): Exploring spatiotemporal patterns of human dynam- ics with geo-targeted social media messages.Big Data & Society, 3(1):2053951716652914, 2016. 3

    2016

  63. [65]

    doi: 10.18653/V1/2023.EMNLP-INDUSTRY.64 2

    doi:10.18653/v1/2023.emnlp-industry.64 2023

  64. [66]

    Zannettou, T

    S. Zannettou, T. Caulfield, E. D. Cristofaro, N. Kourtellis, I. Leontiadis, M. Sirivianos, G. Stringhini, and J. Blackburn. The web centipede: un- derstanding how web communities influence each other through the lens of mainstream and alternative news sources. In S. Uhlig and O. Maennel, eds.,Proceedings of the 2017 Internet Measurement Conference, IMC 20...

    2017

  65. [67]

    Zhang, Y

    W. Zhang, Y . Shen, W. Lu, and Y . Zhuang. Data-copilot: Bridging billions of data and humans with autonomous workflow.CoRR, abs/2306.07209,

    Pith/arXiv arXiv

  66. [68]

    Q. Zhao, M. A. Erdogdu, H. Y . He, A. Rajaraman, and J. Leskovec. SEIS- MIC: A self-exciting point process model for predicting tweet popularity. In L. Cao, C. Zhang, T. Joachims, G. I. Webb, D. D. Margineantu, and G. Williams, eds.,Proceedings of the 21th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Sydney, NSW, Australia, ...

    arXiv 2015

  67. [69]

    Y . Zhao, J. Wang, L. Xiang, X. Zhang, Z. Guo, C. Turkay, Y . Zhang, and S. Chen. Lightva: Lightweight visual analytics with LLM agent- based task planning and execution.CoRR, abs/2411.05651, 2024. doi: 10. 48550/ARXIV.2411.05651 1, 2

    work page internal anchor Pith review Pith/arXiv arXiv doi:10.48550/arxiv.2411.05651 2024

  68. [70]

    L. Yang, T. Sun, M. Zhang, and Q. Mei. We know what @you #tag: does the dual role affect hashtag adoption? In A. Mille, F. Gandon, J. Misselis, M. Rabinovich, and S. Staab, eds.,Proceedings of the 21st World Wide Web Conference 2012, WWW 2012, Lyon, France, April 16-20, 2012, pp. 261–270. ACM, 2012. doi: 10.1145/2187836.2187872 4

    arXiv 2012

  69. [74]

    doi: 10.48550/ARXIV.2306.07209 1, 2

    Pith/arXiv arXiv doi:10.48550/arxiv.2306.07209

  70. [77]

    H. Zhu, X. Yin, J. Ma, and W. Hu. Identifying the main paths of informa- tion diffusion in online social networks.Physica A: Statistical Mechanics and its Applications, 452:320–328, 2016. 3

    2016

  71. [630]

    doi: 10.18653/V1/ 2023.EMNLP-INDUSTRY.59 2

    Association for Computational Linguistics, 2023. doi: 10.18653/V1/ 2023.EMNLP-INDUSTRY.59 2

    doi:10.18653/v1/ 2023

  72. [2012]

    doi: 10.1109/TVCG.2012.291 1, 2, 4

    work page doi:10.1109/tvcg.2012.291 2012

  73. [2017]

    doi: 10.1145/3131365.3131390 4

    arXiv

  74. [2019]

    doi: 10.1145/3343031.3356062 2, 3, 4

    arXiv

  75. [2023]

    doi: 10.1007/S13278-022-01020-5 3, 4

    work page doi:10.1007/s13278-022-01020-5

This paper was first reviewed by deepseek-v4-flash on August 4, 2026.