Pith. sign in

REVIEW 5 major objections 6 minor 72 references

Visibility vs. Engagement: How Two Indian News Websites Reported on LGBTQ+ Individuals and Communities during the Pandemic

T0 review · 5 major / 6 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 During the COVID-19 pandemic, The Times of India and The Indian Express gave LGBTQ+ communities visibility without substantive engagement, and The Times of India's coverage used transphobic language instead of platforming pandemic…

desk verdict A worthwhile empirical snapshot of LGBTQ+ coverage in two Indian papers during COVID, but the paper's sharpest claim—TOI's 'transphobic' language—rests on two examples and needs systematic support. read the letter →

arxiv 2507.15041 v1 pith:KDRQBOGG submitted 2025-07-20 cs.HC

Dhruvee Birla , Nazia Akhtar This is my paper · ORCID
classification cs.HC
keywords LGBTQ+mediarepresentationIndiannewsCOVID-19pandemicsentimentanalysistopicmodellingstancedetectionTheTimesofIndiaExpress
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

The paper's central claim is that during the COVID-19 pandemic, two major Indian English-language news websites, The Times of India (TOI) and The Indian Express (IE), published many articles about LGBTQ+ people but did not engage substantively with their lived realities. The coverage skewed toward Pride celebrations, popular culture, and welfare programs, focused disproportionately on transgender people, and largely omitted pandemic-specific discrimination such as lost livelihoods, family violence, and barriers to healthcare. The stronger conclusion targets TOI, the higher-circulation outlet: its articles were shorter, its language in a manual reading was outdated and transphobic—calling transgender a 'condition' and using the pejorative term 'eunuch'—and the paper concludes it did not provide a substantive platform for LGBTQ+ voices about pandemic discrimination. IE, with more and longer articles, offered more educational coverage, though still not comprehensive. The paper also contributes a method comparison, finding that a fine-tuned domain-specific sentiment model outperforms few-shot ChatGPT-3.5 on LGBTQ+ news headlines.

What carries the argument

The argument is carried by a corpus of 3,037 keyword-filtered articles: 1,576 from the pandemic period (March 2020–August 2021) and 1,461 from a pre-pandemic baseline (January–December 2019). On this corpus the paper applies topic modeling with BERTopic, sentiment analysis comparing a distil-RoBERTa-base model fine-tuned on 1,576 headlines annotated by three LGBTQ+ raters (inter-rater reliability 0.58) against ChatGPT-3.5, and stance detection using the TESTED framework trained on the Fake News Challenge-1 dataset. The interpretive key is pairing those automated results with a manual qualitative reading of articles. The manual layer turns the automated patterns into the paper's central distinction: TOI shows high visibility and positive tone but proves shallow and at times transphobic in language, while IE shows broader and more educative coverage, giving the paper its claim that engagement is not the same as visibility.

What would settle it

A full audit of the 477 TOI pandemic articles—counting 'eunuch' and 'condition' usages and coding how often articles quote LGBTQ+ people on pandemic discrimination, with inter-rater reliability—would decide whether the conclusion that TOI fails to provide a substantive platform is right.

Watch

Extended reading notes

Core claim

The central discovery is that visibility and engagement came apart during the pandemic. The computational results show both outlets concentrated on transgender people and on topics like Pride and popular culture, while sentiment and stance measures showed mostly positive or neutral tones and headlines that agreed with article content. The manual reading is what changes the interpretation: TOI's positive tone coexisted with short articles, thin reporting, and language associated in earlier research with framing trans people as deviant or deceptive. On that basis the paper concludes that the English-language Indian newspaper website TOI does not provide a substantive platform for LGBTQ+ communities to voice their discriminatory experiences during the COVID-19 pandemic in India. The paper further concludes that both outlets lacked depth, that TOI's coverage was the shallower of the two, and that a fine-tuned domain-specific sentiment model (77.5% accuracy) beat few-shot ChatGPT-3.5 (75% accuracy) on this task.

Load-bearing premise

The paper's conclusion that TOI's language is transphobic and obsolete depends on the assumption that the reported examples—calling transgender a 'condition' and using 'eunuch'—are representative of TOI's broader LGBTQ+ coverage, not isolated errors; the paper does not provide a systematic, reliability-checked content analysis of language across the corpus.

Editorial extensions

If this is right

  • If the central claim is right, a news outlet can run a visible pro-LGBTQ+ campaign and still publish reporting that marginalizes the community it claims to support; campaign presence is not evidence of substantive representation.
  • Pandemic coverage that emphasizes Pride and popular culture while under-reporting eviction, unemployment, family violence, and denial of healthcare leaves readers with a distorted picture of how COVID-19 affected LGBTQ+ people in India.
  • Because both outlets focused on transgender people far more than on other LGBQ+ groups, the paper implies that Indian news representation of the community remains narrow even when overall coverage grows.
  • For automated media monitoring, the paper's comparison suggests that domain-specific fine-tuned models should be preferred over general-purpose few-shot LLMs for sentiment analysis of marginalized-group news, at least when accuracy on this kind of text is the goal.
  • The paper finds no major shift in sentiment or stance from before to during the pandemic, so the increased attention did not come with a change in tone.

Reading between the lines

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

  • An implication the authors leave implicit is that the visibility-engagement gap likely generalizes beyond India: article counts and positive sentiment can overstate how well media serve marginalized groups unless language, sourcing, and topic selection are examined qualitatively.
  • The paper's finding that both outlets focused far more on transgender people than on other LGBQ+ groups suggests a testable next question: whether that skew is driven by crime and court stories rather than by welfare and rights stories.
  • Applying the same pipeline to Hindi and other regional-language dailies, which the authors list as future work, could show whether the pattern is specific to English-language news or reflects broader Indian editorial practices.
  • Analyzing reader comments, also flagged as future work, would test whether audience engagement matched the paper's notion of visibility or stayed at the level of sensationalism.
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

5 major / 6 minor

Summary. The paper examines how two Indian English-language newspaper websites, The Times of India (TOI) and The Indian Express (IE), reported on LGBTQ+ individuals and communities during the COVID-19 pandemic (March 2020-August 2021), comparing with a pre-pandemic period (January-December 2019). Using BERTopic topic modeling, distil-RoBERTa and ChatGPT sentiment analysis, TESTED stance classification, and manual qualitative reading of articles, the authors report that coverage focused heavily on transgender communities and on topics such as Pride and popular culture, that TOI articles were shorter than IE articles, and that TOI's language sometimes used transphobic and obsolete terminology. The paper concludes that TOI does not provide a substantive platform for LGBTQ+ communities to voice their discriminatory experiences during the pandemic and that, despite increased awareness, coverage rendered LGBTQ+ communities invisible.

Significance. If the findings were fully supported, the paper would be a useful contribution to scholarship on queer media representation in the Global South, a context that is understudied relative to North America and Europe. The dataset is original, spans a practically important crisis period, and includes a before/during comparison. The use of three LGBTQ+ annotators for the sentiment labels is a notable strength, and the paper honestly discloses some practical difficulties in applying ChatGPT. However, the strongest qualitative claim—that TOI's language is transphobic and obsolete—is currently supported by only two anecdotal examples, and all quantitative comparisons lack statistical testing; the paper's significance therefore depends on substantial additional evidence or careful weakening of the claims.

major comments (5)
  1. [5 Discussion, fourth paragraph; 6 Conclusion] The abstract and conclusion assert that TOI's language was 'transphobic and obsolete' and that 'use of offensive language render[s] LGBTQ+ communities invisible,' but Section 5 supports this with only two article titles ('Eunuch kills lover of over a decade over domestic dispute' and 'Kanpur: Man files FIR against in-laws for misleading him to marry transgender'). The paper does not state how many TOI articles were manually read, how these examples were selected, whether a coding scheme was defined in advance, or whether a second coder independently applied it; the only inter-rater reliability reported (Krippendorff's alpha = 0.58) is for sentiment labels. Without a prevalence estimate, these examples cannot substantiate a claim about TOI's overall reporting style; they are anecdotal. This is load-bearing because the qualitative finding is presented as a key distinction between TOI and IE. The authors should either conduct a systematic content analysis (random sample, predefined codes, multiple coders, prevalence counts) or explicitly downgrade the claim to 'examples of problematic language were found.'
  2. [4 Results, Tables 3–8; 5 Discussion, first paragraph] All quantitative conclusions—article-length differences, sentiment shifts, and stance percentages—are reported without significance tests, confidence intervals, or measures of variability. For example, the claim that TOI's average article length (2563 characters) is 'considerably lesser' than IE's (3909 characters) is presented without standard deviations or a test, and the sentiment percentages in Tables 3–6 and stance percentages in Tables 7–8 may reflect sampling noise rather than stable differences. The paper should add appropriate tests (e.g., bootstrap for means, chi-square or z-tests for proportions) or soften comparative statements accordingly.
  3. [2 Data; 5 Discussion; 6 Conclusion] The pandemic-period corpus comprises all articles matching LGBTQ-related keywords in the date range, not articles specifically about pandemic impacts. Because many retrieved topics are 'American politics' or 'popular culture,' the finding that pandemic-relevant lived experiences were under-covered may be an artifact of the broad keyword filter rather than an editorial priority. The conclusion that TOI 'does not provide a substantive platform for LGBTQ+ communities to voice their discriminatory experiences during the COVID-19 pandemic' should be based on a pandemic-relevant subset or at least an explicit comparison of pandemic-relevant versus non-pandemic-relevant coverage; the current design conflates 'little pandemic coverage' with 'little substantive coverage overall.'
  4. [4 Sentiment Analysis] The sentiment labels used to fine-tune distil-RoBERTa and to prompt ChatGPT were produced by three annotators with Krippendorff's alpha = 0.58, which is below the conventional acceptability threshold of 0.67. The paper neither discusses this low agreement nor treats the labels as noisy; because the models are evaluated against these same labels, the accuracy figures in Section 4 do not establish real-world sentiment detection quality. The authors should either report and discuss the reliability implications or frame the sentiment results as exploratory.
  5. [4 Stance Classification; 5 Sentiment Analysis & Stance Classification] TESTED, trained on FNC-1, assigns labels agree/disagree/discuss/unrelated to the relation between a headline and its article body. The paper's research questions ask about stance toward LGBTQ+ communities, but the reported percentages measure headline-body consistency, not the newspapers' stance toward LGBTQ+ issues. The statement in Section 5 that 'No major shift was noticed in the stance...' therefore does not answer RQ2 as posed. This section should be repositioned as a headline-consistency check, and a target-specific stance measure (or a clear reframing of the claim) is needed.
minor comments (6)
  1. [1 Background] The sentence 'over 13 million vaccines were administered' is likely a scaling error; global COVID-19 vaccination counts are in the billions. Please correct or add the correct reference.
  2. [2 Data, footnote 5] The keyword 'Trans' will match unrelated words such as 'transport,' 'transition,' and 'transmission'; the paper should quantify or filter these false positives to avoid inflating the corpus.
  3. [3 Topic Modelling] BERTopic's UMAP and HDBSCAN hyperparameters (e.g., n_neighbors, min_cluster_size, random seed) are not reported, which prevents reproduction of the topic models.
  4. [3 Sentiment Analysis] Fine-tuning hyperparameters for distil-RoBERTa (learning rate, epochs, batch size, random seed) are not given, despite the Ethics Checklist claiming that all training details were specified.
  5. [4 Results, Tables 3–8] The tables report percentages without the underlying counts or sample sizes, making it impossible to assess the precision of the estimates; please add cell counts or an N per outlet and period.
  6. [Ethics Checklist, item 4(c)] The checklist states that error bars were reported, but I could not find error bars for the fine-tuned model results in Section 4 or the appendix; please align the checklist with the manuscript.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper's claims are empirical descriptions, not derivations that reduce to their inputs.

full rationale

This is an empirical content-analysis paper, not a formal derivation, and no step in its argument reduces to its own inputs by construction. The sentiment classifier is fine-tuned on a labelled training split (1000 headlines) and evaluated on held-out validation and test splits (300 and 276 headlines respectively), so the reported accuracy figures are out-of-sample rather than fitted-to-prediction. The topic-modelling results are descriptive outputs of BERTopic applied to the collected corpus, and the qualitative conclusions about TOI's language are presented as manually identified examples from the dataset, not as quantities derived from those same examples through a fitted model. The closest potential concern is that the 'transphobic and obsolete' language finding rests on a small number of illustrative article titles without reported inter-coder reliability or prevalence statistics; however, that is an evidentiary and generalizability limitation, not a circularity. The paper does not invoke a self-citation as load-bearing for its central claims, and no prediction is equivalent to a fitted parameter by definition. Therefore the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 3 assumptions · 0 invented entities

The paper relies on standard ML models and assumptions about the validity of automated sentiment, topic, and stance measurements. The main free parameters are the undisclosed hyperparameters and prompt settings. No new theoretical entities are introduced.

free parameters (3)
  • Unspecified BERTopic hyperparameters (UMAP n_neighbors, HDBSCAN min_cluster_size, etc.)
    Not reported, so topic stability and cluster assignment cannot be independently assessed.
  • Unspecified fine-tuning hyperparameters for distil-RoBERTa-base (learning rate, epochs, batch size, random seed)
    Not reported, so the final sentiment model cannot be exactly reproduced.
  • ChatGPT prompt and number of few-shot examples
    Not fully specified; the appendix describes re-prompting and accepting all outputs from the rephrased instruction, making results partly dependent on prompt adjustment.
assumptions (3)
  • domain assumption Keyword-based retrieval from the archives captures all relevant LGBTQ+ articles and no irrelevant ones.
    Section 2 defines the filtering; the authors acknowledge in Limitations that the dataset included unrelated election articles.
  • domain assumption The sentiment labels provided by three annotators (Krippendorff's alpha = 0.58) are reliable enough to support comparisons between outlets and periods.
    Section 3 describes the annotation; a moderate alpha of 0.58 is a relatively low level of agreement.
  • domain assumption The TESTED stance model, trained on FNC-1, provides a meaningful measure of the headline-content relationship for this news corpus.
    Section 3 uses this model; the high 'agree' rates suggest the task may be near-trivial for professional news headlines.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Visibility vs. Engagement: How Two Indian News Websites Reported on LGBTQ+ Individuals and Communities during the Pandemic." pith.science (2026). https://pith.science/paper/KDRQBOGG

@misc{pith2026250715041,
  author       = {Pith},
  title        = {Pith review of: Visibility vs. Engagement: How Two Indian News Websites Reported on LGBTQ+ Individuals and Communities during the Pandemic},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/KDRQBOGG}},
  note         = {Machine review of arXiv:2507.15041}
}
read the original abstract

In India, online news media outlets were an important source of information for people with digital access during the COVID-19 pandemic. In India, where "transgender" was legally recognised as a category only in 2014, and same-sex marriages are yet to be legalised, it becomes crucial to analyse whether and how they reported the lived realities of vulnerable LGBTQ+ communities during the pandemic. This study analysed articles from online editions of two English-language newspaper websites, which differed vastly in their circulation figures-The Times of India and The Indian Express. The results of our study suggest that these newspaper websites published articles surrounding various aspects of the lives of LGBTQ+ individuals with a greater focus on transgender communities. However, they lacked quality and depth. Focusing on the period spanning March 2020 to August 2021, we analysed articles using sentiment analysis and topic modelling. We also compared our results to the period before the pandemic (January 2019 - December 2019) to understand the shift in topics, sentiments, and stances across the two newspaper websites. A manual analysis of the articles indicated that the language used in certain articles by The Times of India was transphobic and obsolete. Our study captures the visibility and representation of the LGBTQ+ communities in Indian newspaper websites during the pandemic.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

72 extracted references · 65 canonical work pages

  1. [1]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all mid.sentence after.sentence after.block FUNCTION init.state.consts #0 'before.a...

  2. [2]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 global.max substring 't := if while FUNCTION word.in bbl.in capitalize " " * FUNCT...

  3. [3]

    M.; and Restar, A

    Adamson, T.; Lett, E.; Glick, J.; Garrison-Desany, H. M.; and Restar, A. 2022. Experiences of violence and discrimination among LGBTQ+ individuals during the COVID-19 pandemic: a global cross-sectional analysis. BMJ global health, 7(9): e009400

    work page 2022

  4. [4]

    kerlund, M. 2019. Representations of trans people in Swedish newspapers. Journalism studies, 20(9): 1319--1338

    work page 2019

  5. [5]

    Al-Rawi, A.; Grepin, K.; Li, X.; Morgan, R.; Wenham, C.; and Smith, J. 2021. Investigating public discourses around gender and COVID-19: a social media analysis of Twitter data. Journal of Healthcare Informatics Research, 5: 249--269

    work page 2021

  6. [6]

    Alon, T.; Doepke, M.; Olmstead-Rumsey, J.; and Tertilt, M. 2020. The impact of COVID-19 on gender equality. Technical report, National Bureau of economic research

    work page 2020

  7. [7]

    Aneez, Z.; T Neyazi, A.; Kalogeropoulos, A.; and Nielsen, R. 2019. India digital news report. Reuters

    work page 2019

  8. [8]

    Arakelyan, E.; Arora, A.; and Augenstein, I. 2023. Topic-Guided Sampling For Data-Efficient Multi-Domain Stance Detection. arXiv preprint arXiv:2306.00765

    work page Pith review arXiv 2023

Show all 72 references
  1. [9]

    Arora, S.; and Kumar, K. J. 2023. Media Narratives and the COVID-19 Pandemic: The Asian Experience. Taylor & Francis

    2023

  2. [10]

    Audit Bureau of Circulations . 2023. Highest Circulated Daily Newspapers (Languages Wise). http://www.auditbureau.org/files/JD Accessed: 2024-05-13

    2023

  3. [11]

    Barker-Plummer, B. 2013. Fixing Gwen: News and the mediation of (trans) gender challenges. Feminist Media Studies, 13(4): 710--724

    2013

  4. [12]

    K.; Me s trovi \'c , A.; Beliga, S.; and Martin c i \'c -Ip s i \'c , S

    Bogovi \'c , P. K.; Me s trovi \'c , A.; Beliga, S.; and Martin c i \'c -Ip s i \'c , S. 2021. Topic modelling of Croatian news during COVID-19 pandemic. In 2021 44th International Convention on Information, Communication and Electronic Technology (MIPRO), 1044--1051. IEEE

    2021

  5. [13]

    Brown, T.; Mann, B.; Ryder, N.; Subbiah, M.; Kaplan, J. D.; Dhariwal, P.; Neelakantan, A.; Shyam, P.; Sastry, G.; Askell, A.; Agarwal, S.; Herbert-Voss, A.; Krueger, G.; Henighan, T.; Child, R.; Ramesh, A.; Ziegler, D.; Wu, J.; Winter, C.; Hesse, C.; Chen, M.; Sigler, E.; Litw...

    2020

  6. [14]

    Capuzza, J. C. 2016. Improvements still needed for transgender coverage. Newspaper Research Journal, 37(1): 82--94

    2016

  7. [15]

    Carli, L. L. 2020. Women, gender equality and COVID-19. Gender in management: an International Journal, 35(7/8): 647--655

    2020

  8. [16]

    Chakraborty, A.; Paranjape, B.; Kakarla, S.; and Ganguly, N. 2016. Stop clickbait: Detecting and preventing clickbaits in online news media. In 2016 IEEE/ACM international conference on advances in social networks analysis and mining (ASONAM), 9--16. IEEE

    2016

  9. [17]

    Fairclough, N. 2013. Language and power. Routledge

    2013

  10. [18]

    N.; Salerno, J.; Williams, N

    Fish, J. N.; Salerno, J.; Williams, N. D.; Rinderknecht, R. G.; Drotning, K. J.; Sayer, L.; and Doan, L. 2021. Sexual minority disparities in health and well-being as a consequence of the COVID-19 pandemic differ by sexual identity. LGBT health, 8(4): 263--272

    2021

  11. [19]

    S.; Friedman, J.; Spencer, C

    Flor, L. S.; Friedman, J.; Spencer, C. N.; Cagney, J.; Arrieta, A.; Herbert, M. E.; Stein, C.; Mullany, E. C.; Hon, J.; Patwardhan, V.; et al. 2022. Quantifying the effects of the COVID-19 pandemic on gender equality on health, social, and economic indicators: a comprehensive ...

    2022

  12. [20]

    Ganguly, D.; and Singh, R. 2021. The transgender humanitarian crisis during the Covid-19 pandemic in India. Intersections: Gender and Sexuality in Asia and the Pacific

    2021

  13. [21]

    Gausman, J.; and Langer, A. 2020. Sex and gender disparities in the COVID-19 pandemic. Journal of Women's Health, 29(4): 465--466

    2020

  14. [22]

    Ghasiya, P.; and Okamura, K. 2021. Investigating COVID-19 news across four nations: A topic modeling and sentiment analysis approach. Ieee Access, 9: 36645--36656

    2021

  15. [23]

    Grootendorst, M. 2022. BERTopic: Neural topic modeling with a class-based TF-IDF procedure. arXiv preprint arXiv:2203.05794

    2022 arXiv

  16. [24]

    Gupta, K. 2019. Response and responsibility: Mainstream media and Lucy Meadows in a post-Leveson context. Sexualities, 22(1-2): 31--47

    2019

  17. [25]

    I am Chelsea Manning

    Hackl, A. M.; Becker, A. B.; and Todd, M. E. 2016. “I am Chelsea Manning”: Comparison of gendered representation of Private Manning in US and international news media. Journal of homosexuality, 63(4): 467--486

    2016

  18. [26]

    Hardalov, M.; Arora, A.; Nakov, P.; and Augenstein, I. 2021. Cross-domain label-adaptive stance detection. arXiv preprint arXiv:2104.07467

    2021 arXiv

  19. [27]

    Hindarto, I. H. 2022. Investigating How the National Online Media Reported the LGBT Community During the COVID-19 Pandemic. Profetik: Jurnal Komunikasi, 15(2): 208--227

    2022

  20. [28]

    Hiscott, J.; Alexandridi, M.; Muscolini, M.; Tassone, E.; Palermo, E.; Soultsioti, M.; and Zevini, A. 2020. The global impact of the coronavirus pandemic. Cytokine & growth factor reviews, 53: 1--9

    2020

  21. [29]

    Hu, Krystal . 2023. ChatGPT Sets Record for Fastest-Growing User Base - Analyst Note. http://www.reuters.com/technology/chatgpt-sets-record-fastest-growing-user-base-analyst-note-2023-02-01/. Accessed: 2024-05-13

    2023

  22. [30]

    Jain, J.; Dey, D.; Kelkar, B.; and Ahlawat, K. 2021. Analysis of Indian News with Corona Headlines Classification. In International Conference on Artificial Intelligence and Speech Technology, 116--126. Springer

    2021

  23. [31]

    Jiang, T.; Guo, Q.; Xu, Y.; Zhao, Y.; and Fu, S. 2019. What prompts users to click on news headlines? A clickstream data analysis of the effects of news recency and popularity. In International Conference on Information, 539--546. Springer

    2019

  24. [32]

    Koco \'n , J.; Cichecki, I.; Kaszyca, O.; Kochanek, M.; Szyd o, D.; Baran, J.; Bielaniewicz, J.; Gruza, M.; Janz, A.; Kanclerz, K.; et al. 2023. ChatGPT: Jack of all trades, master of none. Information Fusion, 99: 101861

    2023

  25. [33]

    Konnoth, C. 2020. Supporting LGBT communities in the COVID-19 pandemic. 2020). Assessing Legal Responses to COVID-19. Boston: Public Health Law Watch, U of Colorado Law Legal Studies Research Paper, 20--47

    2020

  26. [34]

    KPMG India . 2020. Synopsis: KPMG India Media and Entertainment Report 2020. https://assets.kpmg.com/content/dam/kpmg/in/pdf/2020/09/synopsis-kpmg-india-media-and-entertainment-2020.pdf. Accessed: 2024-08-25

    2020

  27. [35]

    K \"u c \"u k, D.; and Can, F. 2020. Stance detection: A survey. ACM Computing Surveys (CSUR), 53(1): 1--37

    2020

  28. [36]

    K.; Ponnusamy, R.; Priyadharshini, R.; Buitelaar, P.; and Chakravarthi, B

    Kumaresan, P. K.; Ponnusamy, R.; Priyadharshini, R.; Buitelaar, P.; and Chakravarthi, B. R. 2023. Homophobia and transphobia detection for low-resourced languages in social media comments. Natural Language Processing Journal, 5: 100041

    2023

  29. [37]

    Listiorini, D.; and Vidiadari, I. S. 2022. News of LGBT on online media in 2020: endless stigma. Jurnal Studi Komunikasi, 6(2): 531--546

    2022

  30. [38]

    Liu, Q.; Zheng, Z.; Zheng, J.; Chen, Q.; Liu, G.; Chen, S.; Chu, B.; Zhu, H.; Akinwunmi, B.; Huang, J.; et al. 2020. Health communication through news media during the early stage of the COVID-19 outbreak in China: digital topic modeling approach. Journal of medical Internet r...

    2020

  31. [39]

    Liu, Y.; Ott, M.; Goyal, N.; Du, J.; Joshi, M.; Chen, D.; Levy, O.; Lewis, M.; Zettlemoyer, L.; and Stoyanov, V. 2019. Roberta: A robustly optimized bert pretraining approach. arXiv preprint arXiv:1907.11692

    2019 arXiv

  32. [40]

    J.; Bouchoucha, S

    Lucas, J. J.; Bouchoucha, S. L.; Afrouz, R.; Reed, K.; and Brennan-Olsen, S. L. 2022. LGBTQ+ loss and grief in a cis-heteronormative pandemic: a qualitative evidence synthesis of the COVID-19 literature. Qualitative Health Research, 32(14): 2102--2117

    2022

  33. [41]

    MacKenzie, G.; and Marcel, M. 2009. Media coverage of the murder of US transwomen of color. Local violence, global media: Feminist analyses of gendered representations, 79--106

    2009

  34. [42]

    McQuail, D. 1987. Mass communication theory: An introduction. Sage Publications, Inc

    1987

  35. [43]

    Media Research Users Council , R. 2019. Indian Readership Survey Q4 2019 Highlights. Accessed: 2024-09-05

    2019

  36. [44]

    Mohammad, S.; Kiritchenko, S.; Sobhani, P.; Zhu, X.; and Cherry, C. 2016. Semeval-2016 task 6: Detecting stance in tweets. In Proceedings of the 10th international workshop on semantic evaluation (SemEval-2016), 31--41

    2016

  37. [45]

    A.; Parker, K

    Morrison, M. A.; Parker, K. M.; Sadika, B.; Sameen, D.-E.; and Morrison, T. G. 2021. ‘Newsworthy enough?’: media framing of Canadian LGBTQ persons’ sexual violence experiences. Psychology & Sexuality, 12(1-2): 96--114

    2021

  38. [46]

    T.; and Nielsen, R

    Newman, N.; Fletcher, R.; Schulz, A.; Andi, S.; Robertson, C. T.; and Nielsen, R. K. 2021. Reuters Institute digital news report 2021. Reuters Institute for the study of Journalism

    2021

  39. [47]

    H.; et al

    Organization, W. H.; et al. 2022. WHO Director-General’s opening remarks at the media briefing on COVID-19. January, 30

    2022

  40. [48]

    Palau Sampio, D. 2016. Reference press metamorphosis in the digital context: clickbait and tabloid strategies in Elpais. com. Communication & Society, 29(2)

    2016

  41. [49]

    M.; Wang, X.; Xu, J.; P \'e rez-Bill, E.; Bagnarol, R

    Phillips Ii, G.; Felt, D.; Ruprecht, M. M.; Wang, X.; Xu, J.; P \'e rez-Bill, E.; Bagnarol, R. M.; Roth, J.; Curry, C. W.; and Beach, L. B. 2020. Addressing the disproportionate impacts of the COVID-19 pandemic on sexual and gender minority populations in the United States: ac...

    2020

  42. [50]

    Pomerleau, D.; and Rao, D. 2017. Fake news challenge stage 1 (FNC-I): Stance detection. URL www. fakenewschallenge. org

    2017

  43. [51]

    Qin, C.; Zhang, A.; Zhang, Z.; Chen, J.; Yasunaga, M.; and Yang, D. 2023. Is chatgpt a general-purpose natural language processing task solver? arXiv preprint arXiv:2302.06476

    2023 arXiv

  44. [52]

    Raj, S. 2023. Gender, Media and the Covid-19 Pandemic. Media Narratives and the COVID-19 Pandemic: The Asian Experience

    2023

  45. [53]

    V.; Prates, R.; Kwak, H.; and An, J

    Rieis, J.; de Souza, F.; de Melo, P. V.; Prates, R.; Kwak, H.; and An, J. 2015. Breaking the news: First impressions matter on online news. In Proceedings of the international AAAI conference on web and social media, 357--366

    2015

  46. [54]

    Roy, Raina . 2020. Coronavirus: Kolkata’s Trans Community Has Been Locked out of Healthcare and Livelihood. https://scroll.in/article/968182/coronavirus-kolkatas-trans-community-has-been-locked-out-of-healthcare-and-livelihood. Accessed: 2024-05-14

    2020

  47. [55]

    Saxena, V. K. 2021. Impact of Covid-19 on Indian national print media. International Journal of Research and Analytical Reviews (IJRAR)

    2021

  48. [56]

    clickbait

    Scacco, J. M.; and Muddiman, A. 2016. Investigating the influence of “clickbait” news headlines. Engaging News Project Report

    2016

  49. [57]

    Semykina, K. 2018. Media Construction of LGBT Prides in Russia: Framing Dynamics and Frame Resonance. Higher School of Economics Research Paper No. WP BRP, 81

    2018

  50. [58]

    Shah, C.; Merchant, R.; Mahajan, S.; and Nevatia, S. 2015. No outlaws in the gender galaxy. Zubaan

    2015

  51. [59]

    Stefanov, P.; Darwish, K.; Atanasov, A.; and Nakov, P. 2020. Predicting the topical stance and political leaning of media using tweets. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, 527--537

    2020

  52. [60]

    Stemple, L.; Karegeya, P.; and Gruskin, S. 2016. Human rights, gender, and infectious disease: from HIV/AIDS to Ebola. Hum. Rts. Q., 38: 993

    2016

  53. [61]

    R.; Acic, I.; and Rhea, S

    Stevens, H. R.; Acic, I.; and Rhea, S. 2021. Natural language processing insight into LGBTQ+ youth mental health during the COVID-19 pandemic: longitudinal content analysis of anxiety-provoking topics and trends in emotion in LGBTeens microcommunity subreddit. JMIR public heal...

    2021

  54. [62]

    The Times of India . 2023. Times Out & Proud: It’s time to claim your space with pride. https://timesofindia.indiatimes.com/Times-Out-Proud-Its-time-to-claim-your-space-with-pride/campaignlanding/69161947.cms?from=mdr/. Accessed: 2024-05-14

    2023

  55. [63]

    B.; Tshuma, L

    Tshuma, B. B.; Tshuma, L. A.; and Ndlovu, N. 2022. Media discourses on gender in the time of COVID-19 pandemic in Zimbabwe. Health crises and media discourses in Sub-Saharan Africa, 267

    2022

  56. [64]

    Velasco, Gabi and Langness, Mel . 2020. COVID-19 Action That Centers Black LGBTQ People Can Address Housing Inequities. https://www.urban.org/urban-wire/covid-19-action-centers-black-lgbtq-people-can-address-housing-inequities. Accessed: 2024-05-14

    2020

  57. [65]

    C.; Ghazzai, H.; and Massoud, Y

    Wan, X.; Lucic, M. C.; Ghazzai, H.; and Massoud, Y. 2021. Topic modeling and progression of American digital news media during the onset of the COVID-19 pandemic. IEEE Transactions on Technology and Society, 3(2): 111--120

    2021

  58. [66]

    Wang, Z.; Xie, Q.; Feng, Y.; Ding, Z.; Yang, Z.; and Xia, R. 2023. Is ChatGPT a good sentiment analyzer? A preliminary study. arXiv preprint arXiv:2304.04339

    2023 arXiv

  59. [67]

    E.; Feng, H.; Gr \'e pin, K

    Wenham, C.; Smith, J.; Davies, S. E.; Feng, H.; Gr \'e pin, K. A.; Harman, S.; Herten-Crabb, A.; and Morgan, R. 2020. Women are most affected by pandemics—lessons from past outbreaks. Nature, 583(7815): 194--198

    2020

  60. [68]

    Whittington, C.; Hadfield, K.; and Calder \'o n, C. 2020. The lives & livelihoods of many in the LGBTQ community are at risk amidst COVID-19 crisis. Human Rights Campaign Foundation

    2020

  61. [69]

    Wiktorek, A. E. 2015. Prohibit, tolerate, or prefer: a content analysis of agenda-setting and the LGBT in MSNBC and Fox News. Liberty University

    2015

  62. [70]

    Willox, D. 2003. Branding Teena:(Mis) representations in the media. Sexualities, 6(3-4): 407--425

    2003

  63. [71]

    World Health Organization . 2023. WHO COVID-19 dashboard. https://data.who.int/dashboards/covid19/vaccines?n=c. Accessed: 2024-01-13

    2023

  64. [72]

    Yuan, Y.; Verma, G.; Keller, B.; and Aledavood, T. 2023. Minority stress experienced by LGBTQ online communities during the COVID-19 pandemic. In Proceedings of the International AAAI Conference on Web and Social Media, volume 17, 936--947

    2023

Pith tools

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