pith. sign in

arxiv: 2606.22785 · v1 · pith:KJJX36DUnew · submitted 2026-06-22 · 💻 cs.SI · cs.CL· cs.CR

Cross-National Information Attacks: A Two-Decade Analysis of Troll Behavior in Korea

Jaehong Kim , Hyeonseung Kim , Jiseon Kim , Alice Oh , Thorsten Holz , Wonjae Lee , Meeyoung Cha This is my paper

Pith reviewed 2026-06-26 06:38 UTC · model grok-4.3

classification 💻 cs.SI cs.CLcs.CR
keywords influence operationstroll detectionKorean commentsmoral condemnationcoordinated manipulationpolitical polarizationexplainable AIlongitudinal analysis
0
0 comments X

The pith

Hierarchical model identifies 23,998 Korean accounts using moral condemnation to target domestic politicians in suspected influence operations.

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

The paper develops an explainable machine learning framework to detect suspected foreign trolling in South Korean news comment sections across nearly 20 years. The model classifies comments on foreign origin, moral-emotional framing, and target country while extracting textual evidence for each decision. When run on 112 million comments from 4 million users, it flags 23,998 accounts whose patterns match coordinated manipulation. These accounts favor moral attacks on domestic political figures over direct foreign promotion, and the attacks draw higher engagement that may widen polarization.

Core claim

The hierarchical model classifies comments along three dimensions central to influence campaigns: foreign origin, moral-emotional framing, and target country. It also extracts brief span-level textual evidence. Applied to 112M South Korean news comments, the model identifies 23,998 accounts exhibiting behavior consistent with coordinated manipulation. These accounts predominantly rely on morally condemning rhetoric that receives significantly higher user engagement, with the highest-engagement comments most frequently targeting domestic political figures on both the left and the right.

What carries the argument

The hierarchical classifier that assigns comments to foreign origin, moral-emotional framing, and target country while returning span-level textual evidence for interpretability.

Load-bearing premise

The classifier can accurately identify foreign origin and coordinated manipulation behavior from comment text alone.

What would settle it

Independent verification of account origins or coordination signals, such as through IP logs or external records, for a sample of the 23,998 flagged accounts.

Figures

Figures reproduced from arXiv: 2606.22785 by Alice Oh, Hyeonseung Kim, Jaehong Kim, Jiseon Kim, Meeyoung Cha, Thorsten Holz, Wonjae Lee.

Figure 1
Figure 1. Figure 1: Framework for analyzing suspected troll strategies, where explainable AI connects detection to subsequent strategy [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: A framework for explainable comment-level troll [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Figure 3: Illustrative output from the final model, with rationale-related spans highlighted. Each span corresponds to a predicted [PITH_FULL_IMAGE:figures/full_fig_p006_3.png] view at source ↗
Figure 4
Figure 4. Figure 4: Global feature-importance and SHAP analyses for the user-level troll-detection model. (a) The top five features ranked by mean absolute SHAP value, averaged across 10-fold cross-validation. (b) SHAP summary plots for the same features, showing the distribution and direction of their effects across users. Higher frequency of following ties to known trolls and higher Praising MNS probability shift prediction… view at source ↗
Figure 5
Figure 5. Figure 5: MNS language frequency and diversity by user [PITH_FULL_IMAGE:figures/full_fig_p009_5.png] view at source ↗
Figure 6
Figure 6. Figure 6: Temporal patterns of troll activity from 2006 to 2025. [PITH_FULL_IMAGE:figures/full_fig_p010_6.png] view at source ↗
Figure 7
Figure 7. Figure 7: Like ratio by rhetorical strategy. (a) Distribution of like ratios across strategies with the neutral threshold (like = dislike) shown as a dashed line; Condemning Korea is the only strategy whose mean exceeds this threshold. Error bars denote 95% confidence intervals around the mean. (b) Predicted like ratio from a fractional logit model as a function of the Condemning Korea probability, with user-cluster… view at source ↗
Figure 8
Figure 8. Figure 8: presents the ten most frequent target spans in Con￾demning Korea comments with the highest like ratio (=1). Seven of the top ten targets are political leaders. The most frequently mentioned figures are Moon Jae-in (liberal-leaning former president; 16,651 mentions), Lee Jae-myeong (liberal￾leaning presidential candidate; 13,522 mentions), Yoon Suk￾yeol (conservative president; 9,887 mentions), and Cho Kuk … view at source ↗
Figure 9
Figure 9. Figure 9: Overlap with known trolls by troll group. [PITH_FULL_IMAGE:figures/full_fig_p017_9.png] view at source ↗
Figure 10
Figure 10. Figure 10: Original Korean version of the model output shown in Figure [PITH_FULL_IMAGE:figures/full_fig_p019_10.png] view at source ↗
Figure 11
Figure 11. Figure 11: Like ratio by rhetoric across Korean administrations. Across five administrations, [PITH_FULL_IMAGE:figures/full_fig_p019_11.png] view at source ↗
read the original abstract

Coordinated foreign influence operations pose a growing threat to online platforms, but detecting state-linked troll activity and tracking its evolution remain challenging. This paper presents an explainable machine learning framework for theory-guided detection and longitudinal analysis of suspected trolling within Korean online news comment sections. Our hierarchical model classifies comments along three dimensions central to influence campaigns: foreign origin, moral-emotional framing, and target country. To support explainability, it also extracts brief span-level textual evidence that provides human-interpretable rationales. We apply the approach to 112M South Korean news comments authored by 4M users over nearly 20 years, identifying 23,998 accounts exhibiting behavior consistent with coordinated manipulation. Analyzing these accounts, we find that they predominantly rely on morally condemning rhetoric rather than direct promotion of foreign-aligned narratives; this rhetoric receives significantly higher user engagement. Among the highest-engagement comments, the moral condemnation most frequently targets domestic political figures (e.g., presidents or party leaders) on both the left and the right, potentially amplifying polarization. Our framework supports transparent platform governance through explainable, evidence-based moderation. These observed rhetorical and engagement patterns can inform how platforms and observatories prioritize defenses and intervene before harmful narrative-target combinations achieve widespread reach.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 1 minor

Summary. The paper claims to develop an explainable hierarchical machine learning framework that classifies Korean news comments along three dimensions (foreign origin, moral-emotional framing, and target country) while extracting span-level textual evidence. Applied to 112M comments by 4M users over nearly 20 years, the approach identifies 23,998 accounts exhibiting behavior consistent with coordinated manipulation. Analysis of these accounts finds predominant use of morally condemning rhetoric (rather than direct foreign promotion) that receives higher engagement, with highest-engagement comments most often targeting domestic political figures on both left and right, potentially amplifying polarization.

Significance. If the classifications hold after proper validation, the work would offer a notable contribution through its scale (two-decade longitudinal dataset), emphasis on explainability via span-level rationales, and theory-guided dimensions tailored to influence campaigns. These elements could support more transparent platform moderation and provide falsifiable patterns for future studies on rhetorical strategies in cross-national operations.

major comments (2)
  1. [Abstract] Abstract: The identification of 23,998 accounts as consistent with coordinated manipulation rests entirely on the hierarchical model's first-stage classification of foreign origin from comment text alone. No training details, validation metrics, baseline comparisons, ground-truth sources (e.g., account metadata, known actor lists, or non-text signals), or error analysis are supplied, preventing assessment of whether the classifications support the reported patterns on engagement and polarization.
  2. [Abstract] Abstract: The downstream claims about moral condemnation being the dominant strategy and its higher engagement rest on the same unvalidated foreign-origin stage; if that stage primarily captures domestic stylistic markers correlated with moral-emotional language, the counts and polarization inferences become unreliable.
minor comments (1)
  1. [Abstract] Abstract: The hedging terms ('suspected', 'consistent with') are appropriate but should be paired with an explicit limitations discussion in the methods regarding text-only inference of foreign origin and coordination.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for their careful review and for identifying areas where the abstract requires greater transparency regarding model validation. We address each major comment below with point-by-point responses and indicate the revisions we will make.

read point-by-point responses

  1. Referee: [Abstract] Abstract: The identification of 23,998 accounts as consistent with coordinated manipulation rests entirely on the hierarchical model's first-stage classification of foreign origin from comment text alone. No training details, validation metrics, baseline comparisons, ground-truth sources (e.g., account metadata, known actor lists, or non-text signals), or error analysis are supplied, preventing assessment of whether the classifications support the reported patterns on engagement and polarization.

    Authors: The referee is correct that the abstract does not supply these details. The full manuscript provides them in the Methods section, including the construction of the training set via expert annotation of comment text for foreign-origin indicators drawn from prior studies of Korean influence operations, cross-validation performance, baseline comparisons against standard text classifiers, and error analysis. Ground truth relies on text-based expert labeling rather than account metadata or non-text signals, which were unavailable from the platform. We will revise the abstract to include a concise summary of the first-stage validation metrics and ground-truth approach so readers can assess reliability without consulting the full methods. revision: yes

  2. Referee: [Abstract] Abstract: The downstream claims about moral condemnation being the dominant strategy and its higher engagement rest on the same unvalidated foreign-origin stage; if that stage primarily captures domestic stylistic markers correlated with moral-emotional language, the counts and polarization inferences become unreliable.

    Authors: We acknowledge the risk of stylistic confounding between stages. The model employs theory-guided, stage-specific features: the foreign-origin classifier targets documented non-native phrasing and trolling markers distinct from moral-emotional language, while the second stage operates conditionally but with independent feature extraction. The manuscript reports that moral-condemnation patterns and engagement differences remain consistent in robustness checks. We will add an explicit discussion of this potential confound plus a new robustness analysis comparing moral-emotional distributions in the foreign-origin subset versus a matched domestic sample, thereby strengthening the claim that the observed patterns are not artifacts of the first stage. revision: partial

Circularity Check

0 steps flagged

No circularity; derivation applies external model to independent data

full rationale

The paper describes a hierarchical ML classifier trained on theory-guided dimensions (foreign origin, moral-emotional framing, target country) and applies it to an independent corpus of 112M comments to count accounts and analyze engagement patterns. No equations, self-citations, fitted parameters presented as predictions, or self-definitional steps appear in the provided text that would reduce the account counts or rhetorical findings to the classifier inputs by construction. The central claims are empirical outputs from data application rather than tautological renamings or load-bearing self-citations.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

Only the abstract is available; no model parameters, background assumptions, or new entities are described.

pith-pipeline@v0.9.1-grok · 5775 in / 971 out tokens · 24551 ms · 2026-06-26T06:38:59.350150+00:00 · methodology

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

53 extracted references

  1. [1]

    GPT-4 technical report, 2023

    Josh Achiam, Steven Adler, Sandhini Agarwal, et al. GPT-4 technical report, 2023

    2023

  2. [2]

    Content-based features predict social media influence operations.Science Advances, 2020

    Meysam Alizadeh, Jacob N Shapiro, Cody Buntain, and Joshua A Tucker. Content-based features predict social media influence operations.Science Advances, 2020

    2020

  3. [3]

    Acting the part: Examining information operations within# BlackLivesMatter discourse.Proceedings of the ACM on Human-Computer Interaction, 2018

    Ahmer Arif, Leo Graiden Stewart, and Kate Starbird. Acting the part: Examining information operations within# BlackLivesMatter discourse.Proceedings of the ACM on Human-Computer Interaction, 2018

    2018

  4. [4]

    Bernal, C

    A. Bernal, C. Carter, I. Singh, K. Cao, and O. Madreperla. Cognitive warfare: An attack on truth and thought. Technical report, NATO Innovation Hub and Johns Hopkins University Applied Physics Laboratory, 2020

    2020

  5. [5]

    JSTOR, 2024

    Jude Blanchette, Ryan Hass, and Lily McElwee.Build- ing International Support for Taiwan. JSTOR, 2024

    2024

  6. [6]

    The MAD model of moral contagion: The role of moti- vation, attention, and design in the spread of moralized content online.Perspectives on Psychological Science, 2020

    William J Brady, Molly J Crockett, and Jay J Van Bavel. The MAD model of moral contagion: The role of moti- vation, attention, and design in the spread of moralized content online.Perspectives on Psychological Science, 2020

    2020

  7. [7]

    Emotion shapes the diffu- sion of moralized content in social networks.Proceed- ings of the National Academy of Sciences, 2017

    William J Brady, Julian A Wills, John T Jost, Joshua A Tucker, and Jay J Van Bavel. Emotion shapes the diffu- sion of moralized content in social networks.Proceed- ings of the National Academy of Sciences, 2017

    2017

  8. [8]

    Weaponized health commu- nication: Twitter bots and Russian trolls amplify the vaccine debate.American Journal of Public Health, 2018

    David A Broniatowski, Amelia M Jamison, SiHua Qi, Lulwah AlKulaib, Tao Chen, Adrian Benton, Sandra C Quinn, and Mark Dredze. Weaponized health commu- nication: Twitter bots and Russian trolls amplify the vaccine debate.American Journal of Public Health, 2018

    2018

  9. [9]

    Lan- guage models are few-shot learners

    Tom Brown, Benjamin Mann, Nick Ryder, et al. Lan- guage models are few-shot learners. InProceedings of the 34th International Conference on Neural Informa- tion Processing Systems, 2020

    2020

  10. [10]

    cognitive warfare

    Bernard Claverie and François Du Cluzel. “cognitive warfare”: The advent of the concept of “cognitics” in the field of warfare. InCognitive Warfare: The Future of Cognitive Dominance. NATO Collaboration Support Office, 2022

    2022

  11. [11]

    GoE- motions: A dataset of fine-grained emotions

    Dorottya Demszky, Dana Movshovitz-Attias, Jeongwoo Ko, Alan Cowen, Gaurav Nemade, and Sujith Ravi. GoE- motions: A dataset of fine-grained emotions. InPro- ceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 2020

    2020

  12. [12]

    Span- oriented information extraction: A unified framework

    Yifan Ding, Michael Yankoski, and Tim Weninger. Span- oriented information extraction: A unified framework. ACM SIGKDD Explorations Newsletter, 2025

    2025

  13. [13]

    An anal- ysis of emotions and the prominence of positivity in# BlackLivesMatter tweets.Proceedings of the National Academy of Sciences, 2022

    Anjalie Field, Chan Young Park, Antonio Theophilo, Jamelle Watson-Daniels, and Yulia Tsvetkov. An anal- ysis of emotions and the prominence of positivity in# BlackLivesMatter tweets.Proceedings of the National Academy of Sciences, 2022

    2022

  14. [14]

    ‘fake news’ is the invention of a liar: How false information circulates within the hybrid news sys- tem.Current Sociology, 2019

    Fabio Giglietto, Laura Iannelli, Augusto Valeriani, and Luca Rossi. ‘fake news’ is the invention of a liar: How false information circulates within the hybrid news sys- tem.Current Sociology, 2019

    2019

  15. [15]

    The moral emotions

    Jonathan Haidt. The moral emotions. InHandbook of Affective Sciences. Oxford University Press, 2003

    2003

  16. [16]

    Commenters and lurkers: Navigating the two-step flow of communication in online news dis- course.New Media & Society, 2025

    Jiyoung Han. Commenters and lurkers: Navigating the two-step flow of communication in online news dis- course.New Media & Society, 2025

    2025

  17. [17]

    Hans W. A. Hanley, Deepak Kumar, and Zakir Du- rumeric. Specious sites: Tracking the spread and sway of spurious news stories at scale. InIEEE Symposium on Security and Privacy, 2024

    2024

  18. [18]

    Large language models for propaganda span annota- tion

    Maram Hasanain, Fatema Ahmad, and Firoj Alam. Large language models for propaganda span annota- tion. InFindings of the Association for Computational Linguistics: EMNLP, 2024

    2024

  19. [19]

    The IRA, social me- dia and political polarization in the united states, 2012–

    Philip N Howard, Bharath Ganesh, Dimitra Liotsiou, John Kelly, and Camille François. The IRA, social me- dia and political polarization in the united states, 2012–

    2012

  20. [20]

    Technical report, Project on Computational Pro- paganda, University of Oxford, 2018

    2018

  21. [21]

    How China’s cog- nitive warfare works: a frontline perspective of Taiwan’s anti-disinformation wars.Journal of Global Security Studies, 2022

    Tzu-Chieh Hung and Tzu-Wei Hung. How China’s cog- nitive warfare works: a frontline perspective of Taiwan’s anti-disinformation wars.Journal of Global Security Studies, 2022

    2022

  22. [22]

    Current state of foreign influence operations: Examples of internet and media misuse

    Institute for National Security Strategy (INSS). Current state of foreign influence operations: Examples of internet and media misuse. https://www.inss.re.kr/ en/News/bbs/news_en_view.do?nttId=41037284, 2024

    2024

  23. [23]

    Identi- fying and quantifying coordinated manipulation of up- votes and downvotes in Naver News comments

    Jiwan Jeong, Jeong-han Kang, and Sue Moon. Identi- fying and quantifying coordinated manipulation of up- votes and downvotes in Naver News comments. In Proceedings of the International AAAI Conference on Web and Social Media, 2020

    2020

  24. [24]

    KOLD: Korean offensive language dataset

    Younghoon Jeong, Juhyun Oh, Jongwon Lee, Jaimeen Ahn, Jihyung Moon, Sungjoon Park, and Alice Oh. KOLD: Korean offensive language dataset. InCon- ference on Empirical Methods in Natural Language Processing, 2022

    2022

  25. [25]

    Does transparency in moderation really matter? user behavior after content removal explanations on Reddit.Proceed- ings of the ACM on Human-Computer Interaction, 2019

    Shagun Jhaver, Amy Bruckman, and Eric Gilbert. Does transparency in moderation really matter? user behavior after content removal explanations on Reddit.Proceed- ings of the ACM on Human-Computer Interaction, 2019

    2019

  26. [26]

    Australia, the utility of force and the society-centric battlespace.International Affairs, 2019

    Maryanne Kelton, Michael Sullivan, Emily Bienvenue, and Zac Rogers. Australia, the utility of force and the society-centric battlespace.International Affairs, 2019

    2019

  27. [27]

    How do moral emotions shape political participation? a cross-cultural analysis of online petitions using language models

    Jaehong Kim, Chaeyoon Jeong, Seongchan Park, Meey- oung Cha, and Wonjae Lee. How do moral emotions shape political participation? a cross-cultural analysis of online petitions using language models. InFindings of the Association for Computational Linguistics, 2024

    2024

  28. [28]

    Characterizing and detecting propaganda-spreading accounts on Telegram

    Klim Kireev, Yevhen Mykhno, Carmela Troncoso, and Rebekah Overdorf. Characterizing and detecting propaganda-spreading accounts on Telegram. InPro- ceedings of the 34th USENIX Conference on Security Symposium, 2025

    2025

  29. [29]

    2025 Media Users in Ko- rea (2025 언론수용자조사 )

    Korea Press Foundation. 2025 Media Users in Ko- rea (2025 언론수용자조사 ). Technical report, Ko- rea Press Foundation, Seoul, South Korea, December

    2025

  30. [30]

    Available at https://www.kpf.or.kr/front/ research/consumerDetail.do?seq=600224

  31. [31]

    KcBERT: Korean comments BERT

    Junbum Lee. KcBERT: Korean comments BERT. In Proceedings of the 32nd Annual Conference on Human and Cognitive Language Technology, 2020

    2020

  32. [32]

    KcELECTRA: Korean comments ELEC- TRA

    Junbum Lee. KcELECTRA: Korean comments ELEC- TRA. https://github.com/Beomi/KcELECTRA, 2021

    2021

  33. [33]

    Decoupled weight decay regularization

    Ilya Loshchilov and Frank Hutter. Decoupled weight decay regularization. InInternational Conference on Learning Representations, 2019

    2019

  34. [34]

    Lundberg and Su-In Lee

    Scott M. Lundberg and Su-In Lee. A unified approach to interpreting model predictions. InProceedings of the 31st International Conference on Neural Information Processing Systems, 2017

    2017

  35. [35]

    Tactics, threats & targets: Modeling disinformation and its mitigation

    Muhammad Shujaat Mirza, Labeeba Begum, Liang Niu, Sarah Pardo, Azza Abouzied, Paolo Papotti, and Christina Pöpper. Tactics, threats & targets: Modeling disinformation and its mitigation. InNetwork and Dis- tributed System Security Symposium, 2023

    2023

  36. [36]

    Censored, suspended, shadow- banned: User interpretations of content moderation on social media platforms.New Media & Society, 2018

    Sarah Myers West. Censored, suspended, shadow- banned: User interpretations of content moderation on social media platforms.New Media & Society, 2018

    2018

  37. [37]

    Reuters Institute for the Study of Journalism, 2025

    Nic Newman, Arguedas Ross Arguedas, Craig T Robert- son, Rasmus Kleis Nielsen, and Richard Fletcher.Digi- tal news report 2025. Reuters Institute for the Study of Journalism, 2025

    2025

  38. [38]

    Econometric methods for fractional response variables with an appli- cation to 401 (k) plan participation rates.Journal of Applied Econometrics, 1996

    Leslie E Papke and Jeffrey M Wooldridge. Econometric methods for fractional response variables with an appli- cation to 401 (k) plan participation rates.Journal of Applied Econometrics, 1996

    1996

  39. [39]

    Moral outrage shapes commitments beyond attention: Multimodal moral emotions on YouTube in Korea and the US

    Seongchan Park, Jaehong Kim, Hyeonseung Kim, Hee- jin Bin, Sue Moon, and Wonjae Lee. Moral outrage shapes commitments beyond attention: Multimodal moral emotions on YouTube in Korea and the US. In Proceedings of the ACM Web Conference, 2026

    2026

  40. [40]

    KLUE: Korean language understanding evaluation

    Sungjoon Park, Jihyung Moon, Sungdong Kim, et al. KLUE: Korean language understanding evaluation. In 35th Conference on Neural Information Processing Sys- tems Track on Datasets and Benchmark, 2021

    2021

  41. [41]

    Strategies and vulnerabili- ties of participants in Venezuelan influence operations

    Ruben Recabarren, Bogdan Carbunar, Nestor Hernan- dez, and Ashfaq Ali Shafin. Strategies and vulnerabili- ties of participants in Venezuelan influence operations. In32nd USENIX Security Symposium, 2023

    2023

  42. [42]

    Trollmagnifier: Detecting state- sponsored troll accounts on Reddit

    Mohammad Hammas Saeed, Shiza Ali, Jeremy Black- burn, Emiliano De Cristofaro, Savvas Zannettou, and Gianluca Stringhini. Trollmagnifier: Detecting state- sponsored troll accounts on Reddit. InIEEE symposium on security and privacy, 2022

    2022

  43. [43]

    How malicious AI swarms can threaten democracy.Science, 2026

    Daniel Thilo Schroeder, Meeyoung Cha, Andrea Baronchelli, Nick Bostrom, Nicholas A Christakis, David Garcia, Amit Goldenberg, Yara Kyrychenko, Kevin Leyton-Brown, Nina Lutz, et al. How malicious AI swarms can threaten democracy.Science, 2026

    2026

  44. [44]

    Troll and divide: the language of online polarization

    Almog Simchon, William J Brady, and Jay J Van Bavel. Troll and divide: the language of online polarization. PNAS Nexus, 2022

    2022

  45. [45]

    Disinformation’s spread: bots, trolls and all of us.Nature, 2019

    Kate Starbird. Disinformation’s spread: bots, trolls and all of us.Nature, 2019

    2019

  46. [46]

    How China fights in large-scale combat operations

    Ian Sullivan. How China fights in large-scale combat operations. Technical report, U.S. Army Training and Doctrine Command (TRADOC G-2), 2025

    2025

  47. [47]

    Label Studio: Data labeling software

    Maxim Tkachenko, Mikhail Malyuk, Andrey Hol- manyuk, and Nikolai Liubimov. Label Studio: Data labeling software. https://github.com/ HumanSignal/label-studio, 2020–2025

    2020

  48. [48]

    Senate Select Committee on Intelligence

    U.S. Senate Select Committee on Intelligence. Report on Russian active measures campaigns and interference in the 2016 U.S. election, volume 2: Russia’s use of social media, with additional views. Technical report, United States Senate, 2019

    2016

  49. [49]

    So- cial media and morality.Annual Review of Psychology, 2024

    Jay J Van Bavel, Claire E Robertson, Kareena Del Rosario, Jesper Rasmussen, and Steve Rathje. So- cial media and morality.Annual Review of Psychology, 2024

    2024

  50. [50]

    Council of Europe, 2017

    Claire Wardle and Hossein Derakhshan.Information disorder: Toward an interdisciplinary framework for research and policymaking. Council of Europe, 2017

    2017

  51. [51]

    The comparison of regression vari- ables.Journal of the Royal Statistical Society: Series B (Methodological), 1959

    Evan J Williams. The comparison of regression vari- ables.Journal of the Royal Statistical Society: Series B (Methodological), 1959

    1959

  52. [52]

    SoK: Machine learning for misinformation detection

    Madelyne Xiao and Jonathan Mayer. SoK: Machine learning for misinformation detection. In34th USENIX Security Symposium, 2025

    2025

  53. [53]

    Predicting the type and target of offensive posts in social media

    Marcos Zampieri, Shervin Malmasi, Preslav Nakov, Sara Rosenthal, Noura Farra, and Ritesh Kumar. Predicting the type and target of offensive posts in social media. InProceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 2019. A Troll Dataset Summary We define the 70 a...

    2019