All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
After carefully reviewing the revisions you have made in response to the reviewers' comments, I am pleased to inform you that your manuscript has been accepted for publication in PeerJ Computer Science.
Your efforts to address the reviewers’ suggestions have significantly improved the quality and clarity of the manuscript. The changes you implemented have successfully resolved the concerns raised, and the content now meets the high standards of the journal.
Thank you for your commitment to enhancing the paper. I look forward to seeing the final published version.
[# PeerJ Staff Note - this decision was reviewed and approved by Claudio Ardagna, a PeerJ Section Editor covering this Section #]
no comments
no comment
no comment
The manuscript underwent significant improvements during the extensive review process and is now well-structured to explain the detection of coping strategies, one of the important components in the DRIVE Model, using NLP.
Thank you for submitting your manuscript to PeerJ Computer Science. After careful review, the reviewers have raised some concerns regarding the methodology and experimentation that need to be addressed before we can proceed with the publication.
We kindly request that you revise your manuscript in light of the reviewers' comments and make the necessary adjustments. Please also provide a detailed response letter addressing each of the reviewers' suggestions and observations.
We are confident that, with these revisions, your manuscript will be considered for publication.
Thank you again for your contribution, and we look forward to receiving your revised submission.
Thank you for addressing the issues I have raised and refining your manuscript.
Your manuscript's structure and content are well-written and coherent, according to the research questions, which helps readers better understand your study.
The authors have addressed issues well addressed.
The authors have addressed issues well addressed.
Please break down the results into four subsections, which are directly related to the four research questions. The results section of the current manuscript is a huge chunk and lacks conciseness. Please refer to the lines below, indicating where each subsection can start:
Line 489. Defining framework
Line 535. Identifying Coping Strategies
Line 559. Comparison between in-text coping analysis vs. SA
Line 586. Linking words and text to coping strategies
In addition, I suggest revising the subtitle of the discussion section, “Cultural Difference and Its Effects on Resources and Stressors.” This section discusses the unique characteristics of Kuwaitis' coping strategies, which are informed by their cultural and societal background. However, you only highlight their strategies without comparing them to those of people from other cultural backgrounds, which is not aligned with the subtitle of this section.
Despite your endeavor to modify and correct the typos and broken sentences, I still find some. Please double-check the manuscript thoroughly!
<examples>
Line 51 (coping -> Coping)
Line 89 (The -> the)
Line 95. I am not sure using "....and more" is suitable for scientific and academic manuscripts.
Thank you for submitting your manuscript to PeerJ Computer Science. After careful review, the reviewers have raised some concerns regarding the methodology and experimentation that need to be addressed before we can proceed with the publication.
We kindly request that you revise your manuscript in light of the reviewers' comments and make the necessary adjustments. Please also provide a detailed response letter addressing each of the reviewers' suggestions and observations.
We are confident that, with these revisions, your manuscript will be considered for publication.
Thank you again for your contribution, and we look forward to receiving your revised submission.
This study aimed to propose a DRIVE model-based AI framework to detect individuals’ mental health through text in social media and to conduct a case study identifying individuals’ coping strategies used in social media during the pandemic. Overall, the manuscript has been improved significantly. The revised manuscript is well-organized according to the research questions, making it much easier to follow.
However, it still contains several portions that need further revision:
-Please refine the background in the abstract. It includes too much information, which blurs the significance and purpose of your study. Focus on your main research questions and why they need to be addressed.
-Regarding Figure 1, while you described in the text what the bold characters mean, it would be better to include a note indicating them under Figure 1.
- Please reach out to other reviewers who can read your manuscript more thoroughly. I understand that this manuscript has undergone peer review processes provided in the PeerJ journal, but I can still find broken sentences, (un)capitalized first letters, and incomplete sentences.
-Please make the phrasing of your research questions consistent. For instance, 1. Define…, 2. Identify coping strategies that are used…., 3. Provide evidence….., 4. Indicate the set of words and topics used by people in social media.
-I do not think you need to have introductory sentences at the beginning of the section. For example, you could delete the very first sentence on line 140 that states, “This section provides a brief theoretical background of the proposed framework.” Readers already know what you are going to talk about in that section from the subtitles.
-Line 72. “The paper provides a number of evidence ----.” Please add references and clarify what the paper means.
-The DRVIE model consists of three subcomponents, including demands, resources, and individual differences. Although you defined each component and provided examples of them, you frequently referred to the three components as demands, resources, and coping strategies. Please consider whether coping strategies could be considered a part of individual differences and regarded as one of the personal characteristics or tendencies. If so, please add a description of that.
-You mentioned that your case study focused solely on identifying coping strategies used in social media while leaving other two components (i.e., demands and resources) for future study. However, your response letter mentioned that for the remaining two components, demand and resources, you performed a simple text analysis (bigram and topic modeling) to illustrate how they could be expressed in individuals’ generated text. Please clarify how these relate to simple text analysis.
Please be careful to say positive and negative coping strategies because the effects of using each coping strategy on well-being can vary across studies. For example, avoidant strategies are considered to prevent individuals from effectively dealing with the stressful situation itself and the homeostatic imbalance induced by stressors. However, a growing body of literature suggests that avoidant health behaviors can act on a rewarding stress-response pathway and have a positive short-term impact on restoring psychological and physiological homeostasis. This applies to the classification of coping emojis into positive and negative categories. You need more validation for why you classified them as such.
The authors have addressed the comments raised.
The authors have addressed the comments raised.
The authors have addressed the comments raised.
The authors have addressed the comments raised.
I hope this email finds you well. After a thorough review of your manuscript by the assigned reviewers, I would like to inform you that, while there is potential in your work, several significant concerns have been raised regarding the experimentation and methodology.
The reviewers have pointed out that certain aspects of the experimental setup lack sufficient clarity and justification. In particular, they believe that more detailed explanations and stronger validations are necessary to support your findings. Additionally, methodological improvements have been recommended to ensure the robustness and reliability of the results.
In light of these concerns, we are requesting major revisions to the manuscript. We kindly ask that you carefully address each of the reviewers' comments in your revised submission, providing additional detail and supporting evidence where necessary.
Thank you for another opportunity to review the revised manuscript. The idea of this study, which is developing a new mental health model to detect the use of coping strategies under distress in social media, can significantly contribute to various fields of study. I appreciated that you revised the initial manuscript according to the reviewer’s comments. Addressing the research questions directly was of great help to me (and future readers) in understanding what this study examined. However, you need to revise the manuscript further and improve the quality of your writing according to the following issues.
1. Construction of the manuscript
-Your manuscript should be described more concisely and reconstructed with appropriate subtitles. It was difficult for me to follow your points of discourse and main ideas due to the mislabeled or omitted subtitles. It is still a big chunk of text. Please refer to the index I suggest;
<Suggested index>
1. Background
1.1. Frameworks and techniques for mental health
1.1.1. Drive mental health model
1.1.2. Text-based social media minding techniques
1.2. Application to mental health
1.2.1. DERIVE Model application for the evaluation of mental health status
1.2.2. Text-based social media mining in MH-related applications
1.2.3. Coping strategy analysis and detection in social media
1.3. Drive-COPING observatory framework
2. Method
2.1. Sample
2.2. Analysis procedure
2.3. Analysis contents
3. Result
3.1. Coping strategies used in social media during COVID-19
3.2. Bigrams related to coping
3.3. Comparison of SA and the DRIVE model
4. Discussion
5. Conclusion
2. Clarification of your research questions
-Your newly developed model is driven by the DRIVE model (or theory), which consists of resources, demands, and coping strategies. However, your study only focused on coping strategies except for the other two components described in lines 238-239, page 5, and the review response. (Please check the sentences that are incomplete and broken.) Then, I am wondering if it is appropriate to name your model “DRIVE model-based mental coping.”
You explained the DRIVE model well, including its subcomponents and definitions. It is good to know what this model refers to, but it is confusing that your model considers all three components. Then you can also delete “demands and resources” in research question 1 (Define a machine learning framework that can identify and classify individual demands, resources, and coping strategies from text posted on social media) because you defined and developed the new model considering coping strategies only.
-Figure 1 shows the holistic process of how your model is developed and can be implemented later. Please indicate visually which steps this study focused on (i.e., positive and negative coping strategies).
-Research question 3 (How are people expressing their coping strategies?) needs to be rephrased so that it directly asks what you are testing. The word ‘How’ encompasses various meanings. More directly, you could ask, “What kinds of words were used in social media as a way to cope with covid-related stress?”
You introduced various machine learning approaches that are relevant to your study, such as SA (Vander, TextBlob), EA (extension of SA), and toping modeling (LDA, STP). For readers not in this field of study but interested in mental health and management, it would be great to create a table or appendix that summarizes these machine-learning approaches so that they can easily follow your points.
Please check the use of terminology consistently throughout the text. (e.g., See page 15; TextBlob vs TextBlob SA, VANDER vs VANDER SA)
As mentioned earlier, please separate your findings according to your research questions so that you can highlight your valid findings.
1. Line 18, MH -> Mental Health (MH)
2. Line 150-148
Please reconsider if it is necessary to describe the limitations of the DRIVE model regarding moderation.
3. Line 138, page 3
Is “DRIVE Mental Health Model” the title of the subsection? Then, it should be bold in the text.
4. Line 187-204, page 4.
This paragraph explains the examples of coping strategies, but it seems you just listed several coping strategies without considering their subcategories. Some coping strategies were even described redundantly (e.g., problem-coping strategies and emotion-focused strategies).
5. Line 573, page 22
Please add references for the 0.5 threshold.
6. Check typos/incomplete sentences thoroughly
e.g., Line 177 (i.e., research -> Research), line 238 (i.e., Previous research has consistently This paper is highlighting), line 632 (i.e., coping strategies -> Coping strategies)
The manuscript has been revised to improve the quality of the English language. Its structure conforms to PeerJ guidelines, with separate sections for Introduction, Methods, Results and Discussion. The organisation of the document is consistent and easy to read.
The introduction offers a comprehensive overview of the DRIVE mental health model and explores the significance of coping strategies during the COVID-19 pandemic. It includes an extensive literature review with pertinent citations. The research aims are explicitly defined, emphasizing the development of a framework to identify coping mechanisms in social media content using natural language processing. The data collection process, which utilizes the X social media platform API, is thoroughly explained.
The research question is clear and addresses a significant gap in existing knowledge by moving beyond simple sentiment analysis to focus on the identification of specific coping strategies. The methodology is described with enough detail that the study could be replicated by other researchers. The approach of using a sample of posts that have been manually coded into positive or negative coping categories provides a robust foundation for developing the coping classification models.
The only remaining concern is that the images included in some cases are still of poor quality, but overall the paper can be accepted for publication.
See Additional comments
See Additional comments
See Additional comments
The paper is about social network analysis, which is an interesting topic. The authors propose an AI framework for detecting coping strategies for Mental Health by looking at the emotions expressed by individuals on social networks. The paper is well written and well organized. However, there are several concerns in the current version of the paper that addressing them will increase the quality of this paper.
1 The introduction to the background should be divided into multiple sub-sections so that readers can understand it with clearer logic. In addition, the author should also make it clear what the difference between related work and background is.
2 The quality of Figure 1 needs to be improved. Currently, the structure is not clear and the viewing experience is poor.
3 The same problem occurs with the subsequent figures, which all appear blurry without exception. Please be careful to insert the figure in pdf or eps format.
4 The author's discussion and conclusion sections are both too long. Have you considered expanding the discussion into several subsections and shortening the conclusion?
5 Some related work can be further considered.
[1] Self-Supervised Temporal Graph Learning with Temporal and Structural Intensity Alignment. TNNLS.
[2] A Survey of Knowledge Graph Reasoning on Graph Types: Static, Dynamic, and Multi-Modal. TPAMI.
The manuscript is written clearly and professionally, with unambiguous language used throughout. The authors have provided a strong introduction, outlining the context of the research, the importance of coping mechanisms during COVID-19, and the limitations of traditional sentiment analysis.
The literature is well-referenced and relevant to the topic, providing a solid foundation for the research.
Figures and tables are of high quality, well-labeled, and relevant to the research findings.
Comment: The manuscript could benefit from slight improvements in structuring certain sections to enhance clarity, especially the methods section, which is dense with details.
The research question is clearly defined and addresses an important gap in the current understanding of coping strategies during a global crisis like COVID-19, particularly by using the DRIVE model.
The experimental design is well-planned and executed. The data collection via social media mining and the subsequent classification of coping mechanisms are detailed with sufficient information for replication.
Ethical approval is stated, and the methods follow ethical research standards.
Comment: Ensure that the dataset selection criteria and the rationale for using specific machine learning models (such as XGBoost) are more explicitly justified.
The findings are valid and supported by robust data analysis. The comparison between sentiment analysis and coping-based classification is particularly noteworthy, showing that sentiment analysis can often misclassify coping strategies.
The use of Latent Dirichlet Allocation (LDA) and bigram NLP provides valuable insights into the themes of positive and negative coping strategies.
The conclusions drawn are well-linked to the research question and are consistent with the data presented.
Comment: While the analysis is solid, the limitations of the dataset (such as cultural or geographic biases) should be explicitly addressed.
The paper presents a novel contribution by extending sentiment analysis to coping mechanisms, and the framework is a valuable tool for understanding mental health during crises.
The manuscript could benefit from a slightly more concise discussion section, focusing on key findings and minimizing repetition.
Thank you for submitting your manuscript to PeerJ Computer Science Journal. We have received feedback from the reviewers, and while they have recognized the strengths of your work, they have also provided valuable suggestions for improvement.
Overall, reviewers have provided a positive assessment of your manuscript, praising its clarity, organization, and adherence to PeerJ guidelines. However, there are several areas that require major revisions:
Image Quality: The figures provided are relevant but of poor quality. It's recommended to recreate them with higher resolution for better legibility.
Presentation of Results: While the methods are robust and described well, the presentation of results, especially regarding statistical analyses and classification metrics, needs improvement. More detailed descriptions of results and their implications, along with inclusion of performance metrics like accuracy, precision, recall, and F1-score, would enhance clarity.
Clarity of Conclusions: The conclusions drawn from the results need to be clearer, particularly regarding the implications for public health interventions and mental health support.
Introduction and Justification: Strengthening the introduction by providing explicit justification for the study and expanding on the specific knowledge gap addressed would better contextualize the research within the existing literature.
Language and Grammar: While generally high quality, some phrasing could be improved for clarity.
Supplemental Data: Enhance the supplemental data files with more descriptive metadata identifiers to improve usability for future researchers.
Accessibility of Statistical Analysis: Present the statistical analysis comparing sentiment analysis and coping categories in a more detailed and accessible manner, potentially including relevant test statistics and effect sizes.
Addressing these points in your revision should significantly strengthen your manuscript for resubmission.
no comment. Please refer to the attached letter
no comment. Please refer to the attached letter
no comment. Please refer to the attached letter
no comment. Please refer to the attached letter
The English language used throughout the manuscript is clear, unambiguous, and professional. The writing is of high quality and the terminology is appropriate for the field.
The structure of the manuscript follows the PeerJ guidelines, with clear sections for the Introduction, Methods, Results, and Discussion. The organization is logical and easy to follow.
The raw data has been provided as per the PeerJ policy, which is commendable. This allows for transparency and enables future replication studies.
The introduction provides a thorough background on the DRIVE mental health model and the relevance of coping mechanisms during the COVID-19 pandemic. The literature review is comprehensive and the cited works are highly relevant.
The research objectives are clearly stated, focusing on developing a framework to detect coping mechanisms from social media data using natural language processing techniques.
The data collection methodology, involving the use of the X social media platform API, is well-described.
The natural language processing techniques employed, including sentiment analysis, topic modeling, and classification modeling, are suitable for the research goals.
The figures provided, such as the conceptual framework and example topic modeling results, are relevant but have a relly poor quality. In particular, it is recommended that all images be recreated with higher-quality and more legible content.
The research question is well-defined and addresses an important gap in the literature by going beyond sentiment analysis to focus on the detection of specific coping mechanisms.
The methods are described in sufficient detail to allow for replication. The use of a sample of posts coded into positive or negative coping categories is a robust approach to develop the coping classification models.
The statistical analysis comparing sentiment analysis and coping categories is a valuable addition, though the presentation of these results could be improved for clarity. In particular, it is recommended that the description of the result be expanded to include a more detailed account of the obtained results and their potential implications.
The classification results for the coping inductive model are an important outcome, but more detailed performance metrics (e.g., accuracy, precision, recall, F1-score) would strengthen the presentation of these findings. However, the conclusions drawn from the results are lacking in clarity and the implications for public health interventions and mental health support are not clearly discussed.
The introduction could be further strengthened by providing more explicit justification for the study and expanding on the specific knowledge gap being addressed. This would help to better contextualize the research within the existing literature.
While the language and grammar are generally of high quality, there are a few instances where the phrasing could be improved for clarity.
The supplemental data files would benefit from more descriptive metadata identifiers to enhance their usability for future researchers.
The statistical analysis comparing sentiment analysis and coping categories could be presented in a more detailed and accessible manner, potentially with the inclusion of relevant test statistics and effect sizes.
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.