Dear Prof. Yu,

Thank you very much for submitting your manuscript "Explainable Detection of Adverse Drug Reaction with Imbalanced Data Distribution" for consideration at PLOS Computational Biology.

As with all papers reviewed by the journal, your manuscript was reviewed by members of the editorial board and by several independent reviewers. In light of the reviews (below this email), we would like to invite the resubmission of a significantly-revised version that takes into account the reviewers' comments.

We cannot make any decision about publication until we have seen the revised manuscript and your response to the reviewers' comments. Your revised manuscript is also likely to be sent to reviewers for further evaluation.

When you are ready to resubmit, please upload the following:

[1] A letter containing a detailed list of your responses to the review comments and a description of the changes you have made in the manuscript. Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

[2] Two versions of the revised manuscript: one with either highlights or tracked changes denoting where the text has been changed; the other a clean version (uploaded as the manuscript file).

Important additional instructions are given below your reviewer comments.

Please prepare and submit your revised manuscript within 60 days. If you anticipate any delay, please let us know the expected resubmission date by replying to this email. Please note that revised manuscripts received after the 60-day due date may require evaluation and peer review similar to newly submitted manuscripts.

Thank you again for your submission. We hope that our editorial process has been constructive so far, and we welcome your feedback at any time. Please don't hesitate to contact us if you have any questions or comments.

Sincerely,

Andrey Rzhetsky

Associate Editor

PLOS Computational Biology

Mark Alber

Deputy Editor

PLOS Computational Biology

***********************

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The authors propsed a new machine learning model for Adverse Drug Reaction detection, especially focused on the imbalance issue. They showed promising results based on this model. The reviewer appreciates the novelty of the work as well as the clarity of writing. But there are a few questions that need to be answered before it can be published:

1. While the key component of the proposed model is CRF, the authors have not explained CRF in detail.

2. Could you explain why the proposed model consist of an LSTM followed by the BERT module? To the reviewer's understanding, BERT itself is able to handle the sequential information.

3. Why the train/test splits are different for Tweets and Pubmed data?

4. Fig. 2a and 2c are confusing to the reviewer. Shouldn't the performance converge to a specific value as the epochs increase? If the model is overfitted with large epochs, you should add regularization/dropout to make it converge, instead of using an epoch that is not converged.

5. Equation 12 might be misleading, dash '-' is not distinguishable with minus sign.

Reviewer #2: **** Summary of the paper ****

The objective of this paper is to tackle the data imbalance issue in sequence labeling tasks. Instead of using the conventional softmax in the output layer, a weighted variant of Conditional Random Field (CRF) is proposed to capture the relationship of labels between tokens. Experiments show that the CRF layers improve the existing sequence modeling architectures in two Adverse Drug Reactions (ADR) tasks.

**** Main review ****

Strengths:

(1) I think the data imbalance problem in machine learning in general is interesting. This work focuses specifically on the sequence labeling tasks in Natural Language Processing (NLP). This work might have a greater impact if extended to address the data imbalance problem in other data types such as time-series (e.g., speech).

(2) The authors did a good job in summarizing imbalanced learning techniques in the related works and classified them into data-level methods and algorithm-level methods.

(3) Table 3 in the experiments is extensive. That includes comparison of using softmax or CRF in the output layer of several existing NLP architectures. The experiments did show that the weighted version of CRF slightly improves across 3 metrics (precision, recall and F1-score) on both tasks (Twitter and PubMed).

Weaknesses:

(1) I am not certain if this work fits into the theme of PLOS Computational Biology. Fundamentally, this work is purely NLP. My honest thought is other NLP-focused journals/conferences might be more suitable.

(2) After a quick check in the Internet, I found that the idea of putting CRF on top of existing NLP architectures is not new. To name a few:

- (Huang et al., 2015) Bidirectional LSTM-CRF Models for Sequence Tagging, https://arxiv.org/abs/1508.01991

- (Souza et al., 2019) Portuguese Named Entity Recognition using BERT-CRF, https://arxiv.org/abs/1909.10649

- PyTorch tutorial on a combination of Bi-LSTM and CRF: https://pytorch.org/tutorials/beginner/nlp/advanced_tutorial.html

(3) There is no mentioning that if the implementation/code will be released for reproducibility. What is the choice of deep learning framework to implement the models (e.g., PyTorch or TensorFlow or something else)?

**** Summary of the review ****

The data imbalance problem is meaningful. The results show that models with CRF on top outperform the ones with softmax. However, the novelty of the proposed method is not significant, because many aspects already existed in the literature. My recommendation leans towards rejection.

**** Correctness ****

All of the claims and statements are well-supported and correct.

**** Technical novelty and significance ****

The contributions are only marginally significant or novel. Aspects of the contributions exist in prior work.

**** Empirical novelty and significance ****

The contributions are only marginally significant or novel.

**** Recommendation ****

Marginally below the acceptance threshold. I am more leaning towards rejection.

**** Confidence ****

I am confident in my assessment, but not absolutely certain.

**********

Have the authors made all data and (if applicable) computational code underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data and code 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 and code 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 or code —e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: No: The data used in experiments are publicly available, but the actual implementation/code is not.

**********

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: Yes: Truong Son Hy

Figure 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. 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 us at figures@plos.org.

Data Requirements:

Please note that, as a condition of publication, PLOS' data policy requires that you make available all data used to draw the conclusions outlined in your manuscript. Data must be deposited in an appropriate repository, included within the body of the manuscript, or uploaded as supporting information. This includes all numerical values that were used to generate graphs, histograms etc.. For an example in PLOS Biology see here: http://www.plosbiology.org/article/info%3Adoi%2F10.1371%2Fjournal.pbio.1001908#s5.

Reproducibility:

To enhance the reproducibility of your results, we recommend that you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. Additionally, PLOS ONE offers an option to publish peer-reviewed clinical study protocols. Read more information on sharing protocols at https://plos.org/protocols?utm_medium=editorial-email&utm_source=authorletters&utm_campaign=protocols

Dear Prof. Yu,

We are pleased to inform you that your manuscript 'Explainable Detection of Adverse Drug Reaction with Imbalanced Data Distribution' has been provisionally accepted for publication in PLOS Computational Biology.

Before your manuscript can be formally accepted you will need to complete some formatting changes, which you will receive in a follow up email. A member of our team will be in touch with a set of requests.

Please note that your manuscript will not be scheduled for publication until you have made the required changes, so a swift response is appreciated.

IMPORTANT: The editorial review process is now complete. PLOS will only permit corrections to spelling, formatting or significant scientific errors from this point onwards. Requests for major changes, or any which affect the scientific understanding of your work, will cause delays to the publication date of your manuscript.

Should you, your institution's press office or the journal office choose to press release your paper, you will automatically be opted out of early publication. We ask that you notify us now if you or your institution is planning to press release the article. All press must be co-ordinated with PLOS.

Thank you again for supporting Open Access publishing; we are looking forward to publishing your work in PLOS Computational Biology. 

Best regards,

Andrey Rzhetsky

Associate Editor

PLOS Computational Biology

Mark Alber

Deputy Editor

PLOS Computational Biology

***********************************************************

Reviewer's Responses to Questions

Comments to the Authors:

Please note here if the review is uploaded as an attachment.

Reviewer #1: The previous questions are well answered and the reviewer believes it is ready to publish

**********

Have the authors made all data and (if applicable) computational code underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data and code 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 and code 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 or code —e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

**********

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