Pith. sign in

REVIEW 4 major objections 4 minor 63 references

Investigating Corporate Social Responsibility Initiatives: Examining the case of corporate Covid-19 response

T0 review · 4 major / 4 minor · reviewed 2026-08-09 · deepseek-v4-flash

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

Pith's one-line read This paper argues that standard NLP methods—LDA topic modeling, doc2vec, TF-IDF sentence ranking, and TextRank—can extract the policy-relevant gist from a large corpus of corporate Covid-19 press releases.

desk verdict An honest application of standard NLP methods to Covid-19 CSR press releases, but missing doc2vec tables and lack of summary evaluation keep it from being a validated workflow. read the letter →

arxiv 2502.03421 v1 pith:FZ3B6NGS submitted 2025-02-05 cs.IR

Meheli Basu , Aniruddha Dutta , Purvi Shah This is my paper · ORCID
classification cs.IR
keywords topicmodelingLatentDirichletAllocationdoc2vectextsummarizationRankcorporatesocialresponsibilityCovid-19pressreleasesnaturallanguageprocessing
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

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

The reading

This paper argues that policymakers facing an overload of self-published corporate text can use four widely available NLP tools—LDA topic modeling, doc2vec, TF-IDF sentence ranking, and TextRank—to extract the gist of corporate social responsibility (CSR) activity. The workflow is demonstrated on Covid-19 press releases from the top 15 Nasdaq firms for March through August 2020, and on Amazon's pandemic-response reports. If the claim holds, regulators, media, NGOs, and consumers can monitor what companies say they are doing for employees and communities during a crisis, compare firms and sectors, and set industry-level best-practice expectations. The contribution is a replicable recipe using existing methods rather than a new algorithm.

What carries the argument

The load-bearing machinery is a pipeline pairing two topic models with two summarizers. LDA—a generative model treating each document as a mixture of topics—produces top-keyword lists per period, with a relevance weighting that suppresses corpus-wide frequent words. Doc2vec—a neural model that learns a vector for each document—projects documents into a semantic space, which UMAP dimension reduction and HDBSCAN density-based clustering turn into localized topics, and which allows keyword-based retrieval of supporting sentences. For summarization, TF-IDF sentence ranking scores each sentence by the weight of its words, while TextRank—a graph-based method that ranks sentences by their cosine similarity to other sentences—produces extractive summaries. A central tuning mechanism is the alpha multiplier on the average sentence score in the word-frequency summarizer: lower alpha yields a fuller summary, higher alpha yields a shorter digest.

What would settle it

Compare the topic distributions extracted from firms' press releases with independently audited records of their actual crisis behavior—verified donations, employee-safety complaints, or inspection findings. If firms with weak verified performance produce press-release topics indistinguishable from firms with strong verified performance, the pipeline is recovering messaging rather than engagement, and the policy claim would not hold.

Watch

Extended reading notes

Core claim

The paper's central discovery is that unsupervised topic modeling and extractive summarization, applied to corporate press releases, can surface the main CSR themes firms emphasized during the Covid-19 pandemic. In the early period LDA produced topics labeled prioritize safety, flexible working, accelerated solutions, healthcare efforts, and relief efforts; in the later period the emphasis shifted to helping communities, testing, student learning, and spreading information. Doc2vec adds semantic clustering and keyword-based retrieval of the actual sentences behind a topic. Word-frequency scoring and TextRank then generate readable summaries of a single firm's output, with the summary length controlled by an alpha cutoff. The paper reads these results as evidence that policymakers can run this pipeline on large unstructured text collections to understand corporate societal engagement and guide the standardization of CSR practices.

Load-bearing premise

The analysis stands on the assumption that a firm's press releases are an accurate enough record of what it actually did for employees and communities; if those releases routinely omit failures or exaggerate efforts, the topics and summaries will tell policymakers what companies want them to believe rather than what companies did.

Editorial extensions

If this is right

  • A policymaker can apply the same four-step pipeline to any document corpus—CSR reports, news articles, NGO statements, or social media—and get comparable topic lists and summaries without custom model training.
  • Aggregating by industry makes it possible to see which CSR practices are common and which are neglected, and to standardize best-practice expectations per sector.
  • The gap between self-reported responses and actual conduct, illustrated in the paper by the Tesla example, becomes a governance question that regulators can close only by checking press-release topics against independent evidence.
  • Summary length becomes a policy choice: a smaller alpha gives a fuller picture for deliberation, while a larger alpha gives a concise digest for rapid decisions.
  • Companies themselves can use the methods to benchmark peers' crisis responses and adjust their own CSR strategies and communications.

Reading between the lines

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

  • Beyond the paper, the same workflow could be run on paired corpora—company press releases alongside media and social-media text—so topic models quantify the gap between self-report and independent coverage instead of just acknowledging it.
  • The 2020 topic lists could serve as a crisis-CSR lexicon; reapplying the pipeline during a later crisis would test whether firms' stated priorities shift with public scrutiny.
  • The alpha-tuning knob could be evaluated as a decision variable: future studies could measure whether summaries at different alpha values change stakeholder judgments or policy choices.
  • Because press releases are curated by firms, a natural extension is to apply the identical pipeline to the same firms' mandatory filings and check whether the CSR themes match.
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

4 major / 4 minor

Summary. The paper proposes a workflow for policymakers to extract the 'gist' of large unstructured text corpora using four NLP techniques: LDA and doc2vec for topic modeling, and TF-IDF sentence ranking and TextRank for extractive summarization. It applies these methods to Covid-19-related corporate press releases from the top 15 Nasdaq firms (March-August 2020), reporting LDA topics and coherence scores for two pandemic periods, claiming doc2vec yields localized semantic topics and keyword-based document retrieval, and presenting extractive summaries of Amazon's Covid-19 CSR communications. The paper argues the pipeline is replicable in other social decision-making contexts and can inform CSR standardization and regulation.

Significance. If the workflow were validated, the paper would offer a practical, low-cost template for stakeholders to monitor corporate CSR messaging during crises. Its strengths are the use of a real-world Covid-19 corpus, the coverage of four standard, reproducible methods, the explicit mathematical descriptions, and the honest acknowledgment that self-reported press releases may provide a biased view. However, the evidence in the manuscript is currently insufficient to establish the central claim: the doc2vec results are missing, the summaries are not evaluated against any baseline or reference, and the LDA topic labels are manually assigned without validation. The significance of the paper therefore remains prospective rather than demonstrated.

major comments (4)
  1. [§2.3, Tables 4–5] The doc2vec results are presented as key evidence for the semantic topic-modeling and document-retrieval capabilities, but Tables 4 and 5 are empty in the supplied manuscript. The text reports cosine similarity scores and document IDs (1622 and 1511) without any visible data, so the claims about 'localized topics' and keyword-based retrieval cannot be inspected or replicated. Please provide the actual tables, the full training details (vector dimension, window size, epoch count, HDBSCAN/UMAP parameters), and the exact retrieval procedure used to obtain the scores.
  2. [§2.3, summarization results] The extractive summaries from the TF-IDF sentence-ranking method and TextRank are presented as raw outputs with no reference-based evaluation (e.g., ROUGE), no human assessment, and no baseline such as random sentence selection or lead-3. Without such a baseline, the assertion that these methods 'yield informative summaries' is unsupported; the reader cannot tell whether the summaries are faithful or merely echo frequent words. The choice of alpha = 1.5 and 2 is justified only as yielding 'optimal interpretable summaries,' which is circular unless the optimality criterion is defined. Please add quantitative or human evaluation, or explicitly reframe the results as illustrative examples rather than a validated workflow.
  3. [§2.3, Tables 2–3] The LDA topic labels such as 'Prioritize Safety' and 'Help Community' are manually assigned from keyword lists, but there is no inter-annotator agreement, no validation of the labels, and no confidence interval or baseline for the reported coherence scores. The coherence values alone do not establish that the topics are meaningful, and the cross-period comparison in the Discussion rests on these unvalidated labels. Please provide validation (e.g., multiple annotators, comparison with random topic baselines, or a stated coherence threshold with justification) or temper the interpretive claims.
  4. [Abstract and §2] The abstract claims the steps 'can be replicated,' but the manuscript does not provide the collected corpus, the number of press releases per firm, the scraping and preprocessing protocol (beyond generic NLTK/Gensim references), or any code or data availability statement. Combined with the missing doc2vec tables, this makes the workflow impossible to reproduce as described. Please include a detailed data protocol and make the data and code available, or explicitly state the paper is a conceptual demonstration rather than a fully replicable pipeline.
minor comments (4)
  1. [§2.1.1, Eq. (2)] The parameter alpha in the relevance formula is not the same as the LDA Dirichlet hyperparameter alpha discussed earlier; the text switches between them without distinction, which is confusing. Please rename one of them.
  2. [§2.1.1, Eq. (1)] The notation in Eq. (1) is nonstandard and not referenced later; please align it with standard LDA notation (Blei et al., 2003) or remove it.
  3. [§2.3, table numbering] The text says the top 10 frequent words are 'shown in Table 4' and the top 10 raw sentence scores are 'shown in Table 5,' but Tables 4 and 5 contain doc2vec topics and retrieval, while the word frequencies and scores appear in Tables 6 and 7. This mismatch makes the summarization description hard to follow and should be corrected.
  4. [References and typos] The reference list contains multiple errors: reference 44 is used twice for different works, the in-text citation '[32, 33)' has a misplaced parenthesis, 'Bercelona' should be 'Barcelona,' and Table 3 contains 'June-Aug 20220.' Please proofread the entire manuscript.

Circularity Check

1 steps flagged · score 2.0 of 10

No significant circularity; the only definitional step is the alpha-based summary cutoff, a minor tuning observation rather than a load-bearing prediction.

  1. self definitional [Section 2.3, text summarization results (paragraph following Tables 6 and 7)]
    "An average score of the sentences (threshold) is computed and multiplied with alpha (α) value to determine the sentence cut-off score for qualifying sentences to be part of the final sentence summary. Thus, a sentence gets qualified to appear in the summary if the sentence score is greater than the value threshold * α . We vary the α values and find that α = 1.5, 2 yield optimal interpretable summaries in terms of length."

    The qualifying criterion is sentence score > threshold × α. Therefore, raising α mechanically raises the cut-off and can only remove sentences, never add new ones; the 'finding' that α = 2 produces a shorter summary than α = 1.5 follows algebraically from the defined cutoff. The choice of α = 1.5 and 2 as 'optimal interpretable summaries' is an author judgment based on the generated output rather than an independent evaluation. However, this step only tunes the illustrative Amazon summary and does not support the paper's main claim that standard NLP methods can summarize corporate communications, so the circularity is minor and not load-bearing.

full rationale

The paper's derivation chain is a pipeline demonstration, not a prediction or model-fitting claim: collect corporate press releases, apply standard unsupervised NLP methods (LDA, doc2vec with UMAP/HDBSCAN, TF-IDF sentence scoring, and TextRank), and present the resulting topics and extractive summaries. No core result is derived from a fitted parameter that is then renamed as a finding, and the paper makes no quantitative prediction that could reduce to its inputs by construction. The only definitional step is the alpha cutoff in the word-frequency summarization: because sentences qualify exactly when their score exceeds threshold × α, the observation that larger α gives a shorter summary is a consequence of the formula, not an empirical discovery. This is a minor tuning observation in an illustrative example, not the central claim. There are no load-bearing self-citations, no imported uniqueness theorems, and no renamed known result. Concerns such as the empty Tables 4 and 5, the absence of ROUGE or human evaluation for the summaries, and the reliance on self-reported press releases are validity or reproducibility limitations rather than circularity.

Assumptions & free parameters 6 free parameters · 5 assumptions · 0 invented entities

The paper's conclusions rely on standard algorithms plus several domain choices: the corpus of self-reported press releases, a market-cap based sample, hand-set model parameters, and manual topic labels. None of these are independently validated, and the paper does not release the data needed to check them.

free parameters (6)
  • LDA hyperparameters alpha and beta = not reported
    Chosen as hyperparameters for LDA; exact values are not stated, but they determine topic separation and keyword relevance (Section 2.1.1).
  • Number of LDA topics = 5 per period
    Set to 5 for early and later periods without a data-driven selection procedure (Tables 2 and 3).
  • Doc2vec training settings = min frequency 10, window 10, epochs 10-50
    Hand-selected values reported in Section 2.3; no search or sensitivity analysis is presented.
  • UMAP/HDBSCAN minimum cluster size = 5
    Set in software to 5 for dimensional reduction and clustering (Section 2.3).
  • Summary cut-off alpha = 1.5 and 2
    Chosen because they yield 'optimal interpretable summaries' in terms of length (Section 2.3); output length is directly controlled by this parameter.
  • Custom stopword list = includes Covid-19, Coronavirus
    Words are removed as uninformative, which changes topic distributions (Section 2.1.1).
assumptions (5)
  • standard math Standard LDA generative model, TF-IDF, and TextRank algorithms are correct.
    Invoked in Sections 2.1.1, 2.2.1, and 2.2.2; accepted from cited literature.
  • domain assumption Press releases are a representative or unbiased record of CSR activity.
    The entire analysis uses only company self-reports; the paper acknowledges in the Discussion that other sources would give a more holistic and unbiased view.
  • domain assumption Top 15 Nasdaq firms by market capitalization form a meaningful sample for CSR policymaking.
    Selection is justified only by calling them innovative technology US companies (Section 2), not by policy relevance or representativeness.
  • ad hoc to paper Manual topic labels inferred from keyword lists accurately capture the themes.
    Labels such as 'Healthcare Efforts' and 'Help Community' are assigned by reading keywords (Tables 2 and 3), with no inter-rater or external validation.
  • ad hoc to paper Topic coherence scores above roughly 0.37 indicate useful topics.
    Coherence values are reported but no threshold or comparison is provided (Tables 2 and 3).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Investigating Corporate Social Responsibility Initiatives: Examining the case of corporate Covid-19 response." pith.science (2026). https://pith.science/paper/FZ3B6NGS

@misc{pith2026250203421,
  author       = {Pith},
  title        = {Pith review of: Investigating Corporate Social Responsibility Initiatives: Examining the case of corporate Covid-19 response},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/FZ3B6NGS}},
  note         = {Machine review of arXiv:2502.03421}
}
read the original abstract

In todays age of freely available information, policy makers have to take into account a huge amount of information while making decisions affecting relevant stakeholders. While increase in the amount of information sources and documents increases credibility of decisions based on the corpus of available text, it is challenging for policymakers to make sense of this information. This paper demonstrates how policy makers can implement some of the most popular topic recognition methods, Latent Dirichlet Allocation, Deep Distributed Representation method, text summarization approaches, Word Based Sentence Ranking method and TextRank for sentence extraction method, to sum up the content of large volume of documents to understand the gist of the overload of information. We have applied popular NLP methods to corporate press releases during the early period and advanced period of Covid-19 pandemic which has resulted in a global unprecedented health and socio-economic crisis, when policymaking and regulations have become especially important to standardize corporate practices for employee and social welfare in the face of similar future unseen crises. The steps undertaken in this study can be replicated to yield insights from relevant documents in any other social decision-making context.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

63 extracted references · 61 canonical work pages

  1. [1]

    Sharpe, The Sometime Connection: Public Opinion and Social Policy, Albany, New York: SUNY Press, 1999

    E. Sharpe, The Sometime Connection: Public Opinion and Social Policy, Albany, New York: SUNY Press, 1999

    work page 1999

  2. [2]

    Erikson, M

    R.S. Erikson, M. MacKuen and J. A. Stimson, The Macro Polity, New York: Cambridge University Press, 2002

    work page 2002

  3. [3]

    Soroka, Media, Public Opinion and Foreign Policy, Harvard International Journal of Press and Politics, 8(2003) 27-48

    S.N. Soroka, Media, Public Opinion and Foreign Policy, Harvard International Journal of Press and Politics, 8(2003) 27-48

    work page 2003

  4. [4]

    Eichenberg and R

    R. Eichenberg and R. Stoll, Representing Defence: Democratic Control of the Defence Budget in the United States and Western Europe, Journal of Conflict Resolution, 47(2003) 399-423

    work page 2003

  5. [5]

    Wlezien, Patterns of Representation: Dynamics of Public Preferences and Policy, Journal of Politics, 66(2004)1-24

    C. Wlezien, Patterns of Representation: Dynamics of Public Preferences and Policy, Journal of Politics, 66(2004)1-24

    work page 2004

  6. [6]

    Kotler and N

    P. Kotler and N. Lee, Corporate Social Responsibility: Doing the Most Good for Your Company and Your Cause, John Wiley: Hoboken, NJ, 2005

    work page 2005

  7. [7]

    Škare, and T

    M. Škare, and T. Golja, The impact of government CSR supporting policies on economic growth, Journal of Policy Modeling, 36(2014) 562-577

    work page 2014

  8. [8]

    Kannekanti and V

    S. Kannekanti and V. Muddu, Corporate space for social sciences through corporate social responsibility initiatives rising trend of corporate social responsibility is a boom for Asia's social sciences from theory and practice perspective, Asian Soc. Sci. 4(2008) 149-155

    work page 2008

Show all 63 references
  1. [9]

    O'Rourke, Opportunities and obstacles for corporate social responsibility reporting in developing countries, World Bank, Washington D.C., 2004

    D. O'Rourke, Opportunities and obstacles for corporate social responsibility reporting in developing countries, World Bank, Washington D.C., 2004

    2004

  2. [10]

    Carroll, Corporate Social Responsibility: Evolution of a Definitional Construct, Business & Society, 38(1999) 268-295

    A.B. Carroll, Corporate Social Responsibility: Evolution of a Definitional Construct, Business & Society, 38(1999) 268-295

    1999

  3. [11]

    C. Chen, J. Zhu, J-Y. Yu, and H. Noori, A new methodology for evaluating sustainable product design performance with two-stage network data envelopment analysis, European Journal of Operational Research, 221(2012) 348-359

    2012

  4. [12]

    Duff, Corporate social responsibility reporting in professional accounting firms, The British Accounting Review, 48(2016) 74-86

    A. Duff, Corporate social responsibility reporting in professional accounting firms, The British Accounting Review, 48(2016) 74-86

    2016

  5. [13]

    Hou, Jiran and B.H

    J. Hou, Jiran and B.H. Reber, Dimensions of disclosures: Corporate social responsibility (CSR) reporting by media companies, Public Relations Review, 37(2011) 166-168

    2011

  6. [14]

    M. Morsing, Corporate social responsibility as strategic auto-communication: on the role of external stakeholders for member identification, Business Ethics, the Environment & Responsibility, 15(2006) 171-182

    2006

  7. [15]

    Governance and Accountability Institute. 2015, Seventy-two percent (72%) of the S&P index published corporate sustainability reports in 2013 –dramatically up from 52% in 29 2012 & just about 20% in 2011, 2015, available at: http://www.ga- institute.com/nc/issue-master-system/n...

    2015

  8. [16]

    Crane and G

    A. Crane and G. Sarah, Researching Corporate Social Responsibility Communication: Themes, Opportunities and Challenges, Journal of Management Studies, 53(2016) 1223- 1252

    2016

  9. [17]

    Lehavy, L

    R. Lehavy, L. F and K. Merkley, The effect of annual report readability on analyst following and the properties of their earnings forecasts, The Accounting Review, 86(2011) 1087–1115

    2011

  10. [18]

    Wang, T-S

    Z. Wang, T-S. Hsieh and J. Sarkis (2018), CSR Performance and the Readability of CSR Reports: Too Good to be True?, Corporate Social Responsibility and Environmental Management, 25(2018) 66–79

    2018

  11. [19]

    You and X

    H. You and X. Zhang, Financial reporting complexity and investor underreaction to 10- K information, Review of Accounting Studies, 14(2009) 559–586

    2009

  12. [20]

    2020, COVID-19 pandemic Humanity needs leadership and solidarity to defeat the coronavirus, 2020, available at: https://www1.undp.org/content/brussels/en/home/coronavirus.html

    United Nations Development Program. 2020, COVID-19 pandemic Humanity needs leadership and solidarity to defeat the coronavirus, 2020, available at: https://www1.undp.org/content/brussels/en/home/coronavirus.html

    2020

  13. [21]

    Siddiqui, Faiz, Tesla defied county orders so it could restart production

    F. Siddiqui, Faiz, Tesla defied county orders so it could restart production. Days later, workers tested positive for the coronavirus, The Washington Post, 2020, available at https://www.washingtonpost.com/technology/2020/06/09/tesla-factory-coronavirus/

    2020

  14. [22]

    Aguinis, I

    H. Aguinis, I. Villamor and K.P. Gabriel, Understanding employee responses to COVID-19: A behavioral corporate social responsibility perspective, Management Research, 18(2020) 421–438

    2020

  15. [23]

    Blei, A.Y

    D.M. Blei, A.Y. Ng, and M.I. Jordan, Latent dirichlet allocation, Journal of Machine Learning Research, 3(2003) 993–1022

    2003

  16. [24]

    Misra, O

    H. Misra, O. Cappe and Y. Francois Yvon, Using LDA to detect semantically incoherent documents, Proceedings of the 12th Conference on Computational Natural Language Learning, (2008) 41-48

    2008

  17. [25]

    Donna, U

    G. Donna, U. R. Acharya, R.J. Martis, S.V. Sree, T-C. Lim, T. Ahamed, J.S. Suri, Automated diagnosis of Coronary Artery Disease affected patients using LDA, PCA, ICA and Discrete Wavelet Transform, Knowledge-Based Systems, 37(2013) 274-282

    2013

  18. [26]

    H. Wang, Y. Ding, J. Tang, X. Dong, B. He, J. Qiu, D.J. Wild, Finding Complex Biological Relationships in Recent PubMed Articles Using Bio-LDA, PLoS ONE 6(2011): e17243. https://doi.org/10.1371/journal.pone.0017243

    2011 doi

  19. [27]

    Hofmann, Probabilistic latent semantic indexing, Proceedings of the 22nd Annual International ACM, 1999

    T. Hofmann, Probabilistic latent semantic indexing, Proceedings of the 22nd Annual International ACM, 1999. 30

    1999

  20. [28]

    Sievert and K

    C. Sievert and K. Shirley, LDAvis: A method for visualizing and interpreting topics, Proceedings of the Workshop on Interactive Language Learning, Visualization, and Interfaces, (2014) 63-70

    2014

  21. [29]

    Boyd-Graber, and D.M

    J.L. Boyd-Graber, and D.M. Blei, Syntactic topic models, Advances in neural information processing systems, (2009) 185–192

    2009

  22. [30]

    Syed and M

    S. Syed and M. Spruit, Full-text or abstract? Examining topic coherence scores using latent dirichlet allocation, IEEE International Conference on Data Science and Advanced Analytics (2017) 165–174

    2017

  23. [31]

    K. Yang, Y. Cai, Z. Chen, H. Leung, and R. Lau, Exploring topic discriminating power of words in latent dirichlet allocation, Proceedings of COLING 2016, the 26th International Conference on Computational Linguistics: Technical Papers, (2016) 2238– 2247

    2016

  24. [32]

    Le and Tomas Mikolov, Distributed representations of sentences and documents, ArXiv, abs/1405.4053, 2014

    Quoc V. Le and Tomas Mikolov, Distributed representations of sentences and documents, ArXiv, abs/1405.4053, 2014

    2014 arXiv

  25. [33]

    Radim Reh ˚u ˇ ˇrek and Petr Sojka, Software Framework for Topic Modelling with Large Corpora, Proceedings of the LREC 2010 Workshop on New Challenges for NLP Frameworks, Valletta, Malta, (2010) 45–50

    2010

  26. [34]

    Tomas Mikolov, Kai Chen, Greg Corrado, and Jeffrey Dean, Efficient estimation of word representations in vector space, 2013

    2013

  27. [35]

    Tomas Mikolov, Ilya Sutskever, Kai Chen, Greg S Corrado, and Jeff Dean, Distributed representations of words and phrases and their compositionality, Advances in neural information processing systems, (2013) 3111–3119

    2013

  28. [36]

    context-predicting semantic vectors, Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, Baltimore, Maryland, 1(2014) 238–247

    Marco Baroni, Georgiana Dinu, and Germán Kruszewski, Don’t count, predict! a systematic comparison of context-counting vs. context-predicting semantic vectors, Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics, Baltimore, Maryland, 1(2014) 238–247

    2014

  29. [37]

    Ricardo JGB Campello, Davoud Moulavi, and Jörg Sander, Density-based clustering based on hierarchical density estimates, Pacific-Asia conference on knowledge discovery and data mining, (2013) 160–172

    2013

  30. [38]

    Leland McInnes and John Healy, Accelerated hierarchical density based clustering, IEEE International Conference on Data Mining Workshops (2017)

    2017

  31. [39]

    Leland McInnes, John Healy, and Steve Astels, hdbscan: Hierarchical density based clustering, The Journal of Open Source Software, 11(2017) 205

    2017

  32. [40]

    Leland McInnes, John Healy, and James Melville, Umap: Uniform manifold approximation and projection for dimension reduction, arXiv preprint arXiv:1802.03426, 2018. 31

    2018 arXiv

  33. [41]

    Leland McInnes, John Healy, Nathaniel Saul, and Lukas Grossberger, Umap: Uniform manifold approximation and projection, The Journal of Open Source Software, 3(2018) 86

    2018

  34. [42]

    Laurens van der Maaten and Geoffrey Hinton, Visualizing data using t-sne, Journal of machine learning research, 9(2008) 2579–2605

    2008

  35. [43]

    Geoffrey E Hinton et al., Learning distributed representations of concepts, Proceedings of the eighth annual conference of the cognitive science society, Amherst, MA, 1(1986) 12

    1986

  36. [44]

    Gupta and G

    V. Gupta and G. Lehal, A Survey of Text Summarization Extractive Techniques, Journal of Emerging Technologies in Web Intelligence, 2(2010) 258-268

    2010

  37. [45]

    Jezek and J

    K. Jezek and J. Steinberger, Automatic Text summarization, Vaclav Snasel (Ed.): Znalosti, (2008) 1-12

    2008

  38. [46]

    Kyoomarsi, H

    F. Kyoomarsi, H. Khosravi, E. Eslami and P.K. Dehkordy, Optimizing Text Summarization Based on Fuzzy Logic, Proceedings of Seventh IEEE/ACIS International Conference on Computer and Information Science, IEEE, (2008) 347-352

    2008

  39. [47]

    Chuang and J

    W.T. Chuang and J. Yang, Extracting Sentence Segments for Text Summarization: A Machine Learning Approach, Association for Computing Machinery, (2000) 152-159

    2000

  40. [48]

    Al-Hashemi, Text Summarization Extraction System (TSES) Using Extracted Keywords, International Arab Journal of e-Technology, 1(2010) 164-168

    R. Al-Hashemi, Text Summarization Extraction System (TSES) Using Extracted Keywords, International Arab Journal of e-Technology, 1(2010) 164-168

    2010

  41. [49]

    Luhn, The Automatic Creation of Literature Abstracts, IRE National Convention, New York, (1958) 159-165

    H.P. Luhn, The Automatic Creation of Literature Abstracts, IRE National Convention, New York, (1958) 159-165

    1958

  42. [50]

    Garcia-Herandez and Yulia Ledeneva, Word Sequence Models for Single Text Summarization, IEEE, (2009) 44-48

    R.A. Garcia-Herandez and Yulia Ledeneva, Word Sequence Models for Single Text Summarization, IEEE, (2009) 44-48

    2009

  43. [51]

    Yongzheng, Nur and Evangelos, Narrative Text Classification for Automatic Key Phrase Extraction in Web Document Corpora, WIDM’5, Germany, (2005) 51-57

    2005

  44. [52]

    Kruengkari and C

    V. Kruengkari and C. Jaruskulchai, Generic Text Summarization Using Local and Global Properties of Sentences, Proceedings of the IEEE/WIC international Conference on Web Intelligence (WI’03), 2003

    2003

  45. [53]

    M. Wang, W. Xiaorong and C. Xu, An Approach to Concept Oriented Text Summarization, Proceedings of ISCIT’05, IEEE international conference, China, (2005) 1290-1293

    2005

  46. [54]

    Zamanifar, B

    A. Zamanifar, B. minaei-Bidgoli and M. Sharifi, A New Hybrid Farsi Text Summarization Technique Based on Term Co-Occurrence and Conceptual Property of Text ", Proceedings of Ninth ACIS International Conference on Software Engineering, Artificial Intelligence, Networking and Pa...

    2008

  47. [55]

    Kaikhah, Automatic Text Summarization with Neural Networks, Proceedings of second international Conference on intelligent systems, IEEE, Texas, (2004) 40-44

    K. Kaikhah, Automatic Text Summarization with Neural Networks, Proceedings of second international Conference on intelligent systems, IEEE, Texas, (2004) 40-44

    2004

  48. [56]

    Suanmali, N

    L. Suanmali, N. Salim and B. Mohammed Salem, Fuzzy Logic Based Method for Improving Text Summarization, International Journal of Computer Science and Information Security, 2(2009)

    2009

  49. [57]

    Suanmali, B

    L. Suanmali, B. Mohammed Salem and N. Salim, Sentence Features Fusion for Text summarization using Fuzzy Logic, IEEE, (2009) 142-145

    2009

  50. [58]

    F. Jin, M. Huang and X. Zhu (2009), A Queryspecific Opinion Summarization System, Proceedings of ICCI ’09, 8th IEEE international conference on Cognitive Informatics, Kowloon, Hong Kong, (2009) 428-433

    2009

  51. [59]

    Pembe and T

    F.C. Pembe and T. Güngör, Automated Query biased and Structure-preserving Text Summarization on Web Documents Proceedings of the International Symposium on Innovations in Intelligent Systems and Applications, İstanbul, (2007)

    2007

  52. [60]

    Varadarajan and V

    R. Varadarajan and V. Hristidis, Structure-Based Query-Specific Document Summarization, Proceedings of CIKM’05, ACM, Bremen, Germany, (2005)

    2005

  53. [61]

    Mihalcea and P

    R. Mihalcea and P. Tarau, TextRank: Bringing Order into Text, Proceedings of the 2004 Conference on Empirical Methods in Natural Language Processing, Bercelona, Spain, (2004) 404-411

    2004

  54. [62]

    Barrios, F

    F. Barrios, F. López, L. Argerich and R. Wachenchauzer, Variations of the Similarity Function of TextRank for Automated Summarization, arxiv: 1602.03606, 2016

    2016 arXiv

  55. [63]

    Li and J

    W. Li and J. Zhao, TextRank algorithm by exploiting Wikipedia for short text keywords extraction, 3rd International Conference on Information Science and Control Engineering, (2016) 683-686

    2016

Pith tools

Reviewed August 9, 2026 · model on record in the stance chip above.