PONE-D-20-11657

How many tweets are needed to identify an internet troll?

PLOS ONE

Dear Dr. Monakhov,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

In particular, please address the concerns of Reviewer 2 about the small size of the sample and the generalization of the results.

Please also clarify the difference between trolls and social bots, the details of the tweet sampling and address the other concerns of Reviewer 1.

Please submit your revised manuscript by Jul 20 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Alexandre Bovet, Ph.D.

Academic Editor

PLOS ONE

Journal Requirements:

When submitting your revision, we need you to address these additional requirements.

1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at https://journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and https://journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2. Please amend your title to comply with PLOS publicaiton guidelines: https://journals.plos.org/plosone/s/submission-guidelines#loc-title In this case the title is not sufficiently descriptive of the conducted research.

Additional Editor Comments (if provided):

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Partly

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: This work introduces a linguistic analysis of word co-occurrences and repetitions in social discourse promoted by normal users and internet trolls. The author introduces a simple methodology based on a small number of tweets and leading to a classifier of normal/troll posts with high accuracy.

The manuscript is well written, introducing the reader to the relevant topic of understanding the malicious phenomenon of internet trolling during massive online events and detecting trolls in online platforms. The methodology is simple but novel and results are interesting and well explained. The author should be praised for the innovative combination of simple machine learning techniques, Bayesian inference and sociolinguistic analysis.

There are only a few points, mainly a few missing references and some minor adjustments, that once addressed would improve the quality of this manuscript.

For all the above reasons, I recommend publication of this manuscript in PLoS One with minor revisions.

My points are down below:

• In the Introduction, when the author mentions previous linguistic investigations of trolls’ language, the recent work by Lundberg and Laitinen (2020) might be referenced and briefly discussed. The authors investigated 3.5 million tweets from trolls’ social discourse on Twitter in 2018 but they did not follow a machine learning classification approach as outlined in the current manuscript. Through a simpler frequency analysis, Lundberg and Laitinen also showed that trolls use a smaller linguistic vocabulary, with a sometimes more formal tone. Their results would nicely complement the independent finding presented in this work, on a different dataset and with a different methodology, and could also be interpreted under the light that in order to find more words to express the same concept, not only do trolls mix different contexts but also end up using formal jargon uncommonly used in current language.

• In the Introduction, it is not only trolls that represent a powerful weapon for shifting public opinion but also social bots, fake accounts entirely managed by a software. Do social bots fall into the author’s classification of internet trolls? A recent work by Stella et al. (2019) identified a category of “augmented” human Twitter users that achieve trolling by acting normal but then having their messages re-tweeted or favourited by a large number of social bots.

• In the Results, Figures 1 and 2 are based on sampling 1000 random tweets. How many of these samplings were performed?

• Figures 3 and 4 seem very difficult to parse. Maybe they could be moved to an Appendix. Also, were these networks based on tweets with the same length? The hierarchical edge bundling visualisation seems to suggest that trolls’ linguistic network is more sparse than non-trolls’ but other than that it is of little use here.

• Figure 5-8 seem quite noisy in their left parts. Have they been averaged over 30 different random samplings? As they stand right now, it is really hard to follow the claims from the main text (e.g. two lines “stop mingling after 20 tweets”). Please provide a better labelling, consider changing the scale, smoothing the curves with additional numerical averaging or inserting some insets to highlight specific areas.

• When reporting the accuracy of the classification, could anything be said about what types of errors does the classification perform, e.g. does it tend to identify “false” trolls or “false” normal users?

• About using Trump’s tweets as a reference corpus for the algorithms, it should be noted that a recent work by Clarke and Grieve (2019) pointed out how Trump’s tweets continuously shift between different linguistic paradigms, which would explain why his language does not have high repetition counts like the ones detected in trolls here.

• Maybe “abusive” is not the correct wording for defining Trump’s tweets in a scientific publication without a supporting reference. It is only rather recently that a third party, in this case Twitter, started analysing the semantic content of Trump’s tweets and identified some of them as “potentially misleading” (CNN, 2020). Maybe the last few sentences of the Discussion might be reworded to accommodate specific references outlining how the POTUS’s tweets are controversial, biased and enflaming.

• Typos:

• Line 60 – epy widely?

Reviewer #2: This is an interesting paper that aims to study the differences between tweets posted by congress and state-sponsored trolls.

However, i have serious concerns about the generalizability of the proposed methods.

First, the paper's analysis and results are based on a very small sample of the whole dataset (like 200 or 1000 at some cases tweets out of 1M tweets). This raises concerns on how representative the results are and whether the same results hold if the analysis is done on the whole dataset. I suggest to the authors to expand their analysis so that it includes all the tweets from their dataset rather than sampling a very small percentage of the tweets.

Second, the authors claim that their approach can be used for other languages as well. I think, since each language its different and each type of state-sponsored trolls might employ different tactics, this is a somewhat unsubstantiated claim. I suggest to the authors to expand their analysis to include data from other languages and types of state-sponsored trolls so that we can better assess the generalizability of the proposed methods.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

PONE-D-20-11657R1

Early detection of internet trolls: Introducing an algorithm based on word pairs / single words multiple repetition ratio

PLOS ONE

Dear Dr. Monakhov,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We thank you for the revisions that both reviewers appreciated. Before acceptance, we ask you to address the last issue raised by Reviewer 1. Namely, replicating the analysis of Table 7 for 100, 250 tweets and, if possible, 1000 tweets.

Please submit your revised manuscript by Aug 14 2020 11:59PM. If you will need more time than this to complete your revisions, please reply to this message or contact the journal office at plosone@plos.org. When you're ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

Please include the following items when submitting your revised manuscript:

• A rebuttal letter that responds to each point raised by the academic editor and reviewer(s). You should upload this letter as a separate file labeled 'Response to Reviewers'.
• A marked-up copy of your manuscript that highlights changes made to the original version. You should upload this as a separate file labeled 'Revised Manuscript with Track Changes'.
• An unmarked version of your revised paper without tracked changes. You should upload this as a separate file labeled 'Manuscript'.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter. Guidelines for resubmitting your figure files are available below the reviewer comments at the end of this letter.

If applicable, we recommend that you deposit your laboratory protocols in protocols.io to enhance the reproducibility of your results. Protocols.io assigns your protocol its own identifier (DOI) so that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

We look forward to receiving your revised manuscript.

Kind regards,

Alexandre Bovet, Ph.D.

Academic Editor

PLOS ONE

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. If the authors have adequately addressed your comments raised in a previous round of review and you feel that this manuscript is now acceptable for publication, you may indicate that here to bypass the “Comments to the Author” section, enter your conflict of interest statement in the “Confidential to Editor” section, and submit your "Accept" recommendation.

Reviewer #1: All comments have been addressed

Reviewer #2: All comments have been addressed

**********

2. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

**********

3. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

**********

4. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

**********

5. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

**********

6. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: The author took great care in addressing all my comments and I am positively impressed with the final quality of the manuscript.

I recommend acceptance of the manuscript. I only have one final suggestion about line 424 "First, an anonymous reviewer expressed concern about how representative our results are given the small sample size...". I am confident that any reader could get the same doubt so the reference to the specific reviewer here could be changed. Also, in order to further dissipate these doubts and the risk of overfitting, I would try to replicate the analysis on the new Table 7 for 100 tweets (which is still a relatively small number) also in case of 250 tweets (which becomes more of a diary/storyline) and, if possible, 1000 tweets.

Consider that results cannot be monotonously impeccable as the higher the number of tweets used in the analysis the larger the linguistic scope or set of topics identified in the analysis and thus, because of such larger variability, the accuracy of the model could actually decrease rather than increase. I think this is an important reality check in order to either support or change the authors' claim about the model being "impeccable". If the model kept performing very well also for larger sizes of tweets, this would mean the important possibility of using this methodology not only on Twitter but also on platforms with longer texts like Facebook, a possibility that would be advantageous for the author and his very interesting work.

It is also remarkable that the methodology works also cross-linguistically in Russian, as shown by the author.

I will let the Editor decide on this final point, whether to accept the paper as is or request this minor adjustment. I repeat I was satisfied with the author's work.

Reviewer #2: I would like to thank the authors for addressing the reviewers' comments and resubmitting their work. All the reviewers' comments were addressed in the revised manuscript and the manuscript is substantially stronger now. I recommend acceptance of this work.

**********

7. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: No

Reviewer #2: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email PLOS at figures@plos.org. Please note that Supporting Information files do not need this step.

Early detection of internet trolls: Introducing an algorithm based on word pairs / single words multiple repetition ratio

PONE-D-20-11657R2

Dear Dr. Monakhov,

We’re pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it meets all outstanding technical requirements.

Within one week, you’ll receive an e-mail detailing the required amendments. When these have been addressed, you’ll receive a formal acceptance letter and your manuscript will be scheduled for publication.

An invoice for payment will follow shortly after the formal acceptance. To ensure an efficient process, please log into Editorial Manager at http://www.editorialmanager.com/pone/, click the 'Update My Information' link at the top of the page, and double check that your user information is up-to-date. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to help maximize its impact. If they’ll be preparing press materials, please inform our press team as soon as possible -- no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

Kind regards,

Alexandre Bovet, Ph.D.

Academic Editor

PLOS ONE

Additional Editor Comments:

When preparing the final version, please take into account the following comments:

- Please mention the limitations of your method in the introduction and conclusion. In particular that you only tested you method against tweets of public figures (congresspeople and Trump).

- Please define what is a clique (line 177) in graph theory.

- Rename "Appendix" to "Supporting information" (see https://journals.plos.org/plosone/s/submission-guidelines#loc-manuscript-organization)