Classification aware neural topic model for COVID-19 disinformation categorisation

Xingyi Song; Johann Petrak; Ye Jiang; Iknoor Singh; Diana Maynard; Kalina Bontcheva

doi:10.1371/journal.pone.0247086

Peer Review History

Original SubmissionNovember 4, 2020
4 Dec 2020 Decision Letter - Sanda Martinčić-Ipšić, Editor PONE-D-20-33400 Classification Aware Neural Topic Model for COVID-19 Disinformation Categorisation PLOS ONE Dear Dr. Song, 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. Please submit your revised manuscript by Jan 16 2021 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, Sanda Martinčić-Ipšić, PhD 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.In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability. Upon re-submitting your revised manuscript, please upload your study’s minimal underlying data set as either Supporting Information files or to a stable, public repository and include the relevant URLs, DOIs, or accession numbers within your revised cover letter. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories. Any potentially identifying patient information must be fully anonymized. Important: If there are ethical or legal restrictions to sharing your data publicly, please explain these restrictions in detail. Please see our guidelines for more information on what we consider unacceptable restrictions to publicly sharing data: http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions. Note that it is not acceptable for the authors to be the sole named individuals responsible for ensuring data access. We will update your Data Availability statement to reflect the information you provide in your cover letter. 3.Thank you for submitting the above manuscript to PLOS ONE. During our internal evaluation of the manuscript, we found significant text overlap between your submission and the following previously published work, of which you are an author: http://eprints.whiterose.ac.uk/164746/ We would like to make you aware that copying extracts from previous publications, especially outside the methods section, word-for-word is unacceptable. In addition, the reproduction of text from published reports has implications for the copyright that may apply to the publications. Please revise the manuscript to rephrase the duplicated text, cite your sources, and provide details as to how the current manuscript advances on previous work. Please note that further consideration is dependent on the submission of a manuscript that addresses these concerns about the overlap in text with published work. We will carefully review your manuscript upon resubmission, so please ensure that your revision is thorough. [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: Yes ******** 2. Has the statistical analysis been performed appropriately and rigorously? Reviewer #1: N/A 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: No ****** 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: No ****** 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: Authors have compiled a COVID-19 related dataset and labelled it with 10 categories. They manually annotated a part of data with a good enough Kappa score. To automatically categorize data they used BERT model with a combination of encoder-decoder networks. Also the input for BERT is word-piece-based (as standard for BERT dictionary) while for E-D nework is simple BOW representation. At the end, authors compare and evaluate their results. The paper is well structured, well written. Most of the work is justified and also data and DEMO version of the deployed algorithm is available (full after the publication) online. I have no major comments but I propose authors to update the following parts: - In the beginning of Chapter 4 authors should define the D_{KL}. - In the Figure 1 authors can ommit some basic explanations (e.g. linear layer) as it is supposed the reader already understands BERT model which i more complex. Also in the figure not the same words are used. - In the results I am not sure what is NVDMb and NVDMo? Also, as accuracy is reported, the percentage of majority class should be mentioned (otherwise reader needs to calculate it from Table 4?). - Authors will publish annotated data. Could authors also publish automatically annotated data by their algorithm or publish their code. That would be useful for reproducibility and further comparitons. - Authors sometimes start sentence with a formula (e.g. "p(x\|z) is the generation"....) or reference (e.g. "[8] introduce a"...). I propose to reformat sentences in a way that they do not start like these. Reviewer #2: In the manuscript "Classification Aware Neural Topic Model for COVID-19 Disinformation Categorisation" authors perform topic modelling of COVID-19 disinformation using neural networks. They combine BERT model with Variational Autoencoder and define a CANTM model for a topic generation. The proposed model is evaluated in terms of standard evaluation measures (accuracy, macro F- score and perplexity). The reported results show that the proposed model outperforms some other state-of-the-art approaches and human annotators. The evaluation procedure seems to be correctly implemented. However, the whole manuscript is too extensively written and certain parts of the described approach need to be clarified by explaining the experiment in a more concise text. In general, this research is interesting and valuable, although the rest of the manuscript is not easy to follow. Overall, the manuscript has certain shortcomings, which need to be improved before the work is good enough to be recommended for publication. My suggestions and comments are as follows. 1. Abstract should be rewritten. Now it seems to be slightly misleading because it is written that this research will develop “computational methods to support research on COVID-19 disinformation debunking and its social impact”. In the abstract, it should be emphasized that the main focus of their research is to identify the topic of fake news, not to identify fake news. Furthermore, this abstract is missing an overview of research method and insight into the results. 2. In the introductory section authors describe the motivation, main goals and challenges of their research- The scientific contributions are clearly stated as well. My suggestion is to add one paragraph with concise descriptions of all experiments. 3. Section Dataset Structure is written with too many details. The first part of the Section related to Table 1, together with this table can be moved to the Supplementary materials and leaving only data about dataset statistics. 4. Furthermore, this second Section about the data structure can be a subsection of the section which describes the experiment. 5. The third Section about disinformation category labelling is also too extensive. 6. Section about related work needs to be extended with more references that are relevant for this research. I suggest the authors to include more publications that use BERT model in similar NLP tasks. ****** 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: Yes:** Slavko Žitnik 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. https://doi.org/10.1371/journal.pone.0247086.r001
Revision 1
18 Jan 2021 Author Response Dear Reviewers and AE, We would like to express our appreciation for your insightful comments, constructive suggestions and valuable time for the paper. We sincerely appreciate the encouragement in your comments. We are happy to report that we have addressed the issues that the reviewers raised to the best of our capacity. We have revised the structure and addressed the editorial problems addressed from the reviewers’ suggestions. A color marked-up copy of our manuscript that highlights changes made to the original version is included. Red marks indicate major changes/revisions of paragraphs. Orange marks indicate minor grammatical/structure changes. Blue marks indicate new added paragraphs. The responses to each comment in detail are as follows. Response Reviewer #1: 1. In the beginning of Chapter 4 authors should define the D_{KL}. We added the definition of D_{KL} in Line 141 as “Kullback–Leibler divergence” 2. In the Figure 1 authors can ommit some basic explanations (e.g. linear layer) as it is supposed the reader already understands BERT model which i more complex. Also in the figure not the same words are used. We added the definition of linear layer in Equation 2 and Line 190. We also updated the figure and equations to have consistent wording. 3. In the results I am not sure what is NVDMb and NVDMo? Also, as accuracy is reported, the percentage of majority class should be mentioned (otherwise reader needs to calculate it from Table 4?). We have moved the details of experiment settings from supplemental material to the main content. The definition of NVDMb and NVDMo now in the Line 234- 236, which is “1) original NVDM as described in [8] (“NVDMo” in the results ); 2) NVDM with BERT representation (“NVDMb” in the results).” 4. Authors will publish annotated data. Could authors also publish automatically annotated data by their algorithm or publish their code. That would be useful for reproducibility and further comparitons. We will publish the source code and the data used in this paper. Links to the code and data are added in Section 9 Software and Data 5. Authors sometimes start sentence with a formula (e.g. "p(x\|z) is the generation"....) or reference (e.g. "[8] introduce a"...). I propose to reformat sentences in a way that they do not start like these. Thanks for the suggestion, we have revised the manuscript accordingly Response Reviewer #2: 1. Abstract should be rewritten. Now it seems to be slightly misleading because it is written that this research will develop “computational methods to support research on COVID-19 disinformation debunking and its social impact”. In the abstract, it should be emphasized that the main focus of their research is to identify the topic of fake news, not to identify fake news. Furthermore, this abstract is missing an overview of research method and insight into the results. Thanks for your suggestion, we have rewritten the abstract and parts of the introduction based on your suggestion. 2. In the introductory section authors describe the motivation, main goals and challenges of their research- The scientific contributions are clearly stated as well. My suggestion is to add one paragraph with concise descriptions of all experiments. A concise description of experiments is added in Line 59-64 3. Section Dataset Structure is written with too many details. The first part of the Section related to Table 1, together with this table can be moved to the Supplementary materials and leaving only data about dataset statistics. Thanks for your suggestion, we have now significantly reduced the content of Section Dataset Structure. We left Table 1 (with reduced rows) in the main content, because dataset building is one of the main contributions of this work, and the table is a clear way to present the structure of the data. 4. Furthermore, this second Section about the data structure can be a subsection of the section which describes the experiment. Thanks for your suggestion, we now merged this section with Section 5 “COVID-19 Disinformation Analysis and Discussion” 5. The third Section about disinformation category labelling is also too extensive. Thanks for your suggestion. We have revised the context and merged this section with Section 2 and renamed the sections as “Dataset and Annotation”. Data annotation and the cleaning process are one of the most important steps in our work. This ensures the correctness of the category labeling work and affects the correctness of our experiment and analysis process described in the later sections. We described this process within one page in the current version of the manuscript. 6. Section about related work needs to be extended with more references that are relevant for this research. I suggest the authors to include more publications that use BERT model in similar NLP tasks. Thanks for the suggestion, we now enriched the related work section, and highlighted some of the work using BERT model. Response editor Sanda Martinčić-Ipšić’s queries: 1. Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. We are using the PLOS ONE official Latex template and ensured the manuscript meets PLOS ONE's style requirements 2. In your Data Availability statement, you have not specified where the minimal data set underlying the results described in your manuscript can be found. PLOS defines a study's minimal data set as the underlying data used to reach the conclusions drawn in the manuscript and any additional data required to replicate the reported study findings in their entirety. All PLOS journals require that the minimal data set be made fully available. For more information about our data policy, please see http://journals.plos.org/plosone/s/data-availability. The full dataset is publicly available at: www.kaggle.com/dataset/fd97cd3b8f9b10c1600fd7bbb843a5c70d4c934ed83e74085c50b78d3db18443 The source code is publicly available at: https://github.com/GateNLP/CANTM We added one more section “Section 9 Software and data” in the updated manuscript reporting the availability of source code and dataset. 3. During our internal evaluation of the manuscript, we found significant text overlap between your submission and the following previously published work, of which you are an author: http://eprints.whiterose.ac.uk/164746/ The work http://eprints.whiterose.ac.uk/164746/ is a copy of our arxiv version of the paper (please note the link to the paper is arxiv). White Rose Research Online is a repository that automatically collects research outputs from University of Sheffield and two other universities. The collection includes an arxiv preprint. We deeply appreciate your time reviewing our manuscript, Thank you and best regards. Xingyi Song Attachments Attachment Submitted filename: Response to Reviewers.pdf https://doi.org/10.1371/journal.pone.0247086.r002
2 Feb 2021 Decision Letter - Sanda Martinčić-Ipšić, Editor Classification Aware Neural Topic Model for COVID-19 Disinformation Categorisation PONE-D-20-33400R1 Dear Dr. Song, 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, Sanda Martinčić-Ipšić, PhD Academic Editor PLOS ONE Additional Editor Comments (optional): This manuscript relates to the ongoing outbreak of coronavirus. Given this, I checked the revision, your response to reviewers and data and SW availability. I am glad that the current manuscript revision has addressed all issues adequately and meets required PlosONE criteria. Reviewers' comments: https://doi.org/10.1371/journal.pone.0247086.r003
Formally Accepted
4 Feb 2021 Acceptance Letter - Sanda Martinčić-Ipšić, Editor PONE-D-20-33400R1 Classification Aware Neural Topic Model for COVID-19 Disinformation Categorisation Dear Dr. Song: I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department. If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. 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. If we can help with anything else, please email us at plosone@plos.org. Thank you for submitting your work to PLOS ONE and supporting open access. Kind regards, PLOS ONE Editorial Office Staff on behalf of Dr. Sanda Martinčić-Ipšić Academic Editor PLOS ONE https://doi.org/10.1371/journal.pone.0247086.r004

Open letter on the publication of peer review reports

PLOS recognizes the benefits of transparency in the peer review process. Therefore, we enable the publication of all of the content of peer review and author responses alongside final, published articles. Reviewers remain anonymous, unless they choose to reveal their names.

We encourage other journals to join us in this initiative. We hope that our action inspires the community, including researchers, research funders, and research institutions, to recognize the benefits of published peer review reports for all parts of the research system.

Learn more at ASAPbio .