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.
Dear authors, we are pleased to verify that you meet the reviewer's valuable feedback to improve your research.
Thank you for considering PeerJ Computer Science and submitting your work.
Kind regards
PCoelho
[# PeerJ Staff Note - this decision was reviewed and approved by Yilun Shang, a PeerJ Section Editor covering this Section #]
-
-
-
**PeerJ Staff Note:** Please ensure that all review, editorial, and staff comments are addressed in a response letter and that any edits or clarifications mentioned in the letter are also inserted into the revised manuscript where appropriate.
**Language Note:** The review process has identified that the English language must be improved. PeerJ can provide language editing services - please contact us at [email protected] for pricing (be sure to provide your manuscript number and title). Alternatively, you should make your own arrangements to improve the language quality and provide details in your response letter. – PeerJ Staff
(1) Scenario testing is not clear,
(2) The conclusion lacks sufficient quantitative explanation to support the findings effectively.
-
-
This paper compares deep learning models used to detect false data injection in POS. It is an interesting research topic.
Many experiments have been done.
The comparison results have some values.
1. What is the main innovation of this paper? Authors should further clarify it.
2. Some other deep learning models, such as graph neural networks, should be reviewed and compared.
3. This paper uses the oversampling method to handle the imbalanced dataset problem. Why did not use another method of handling this problem? How about the experimental results using other methods? Authors should supplement these experiments.
-
-
-
This manuscript reports some common False Data Injection Attacks (FDIA) methods targeting Point of Sale (POS) systems. It also discusses the use of deep learning models fine-tuned through specific architectures, such as Random Search Artificial Neural Network (RS-ANN), Bayesian Optimized Convolutional Neural Network Long Short-Term Memory (BO-CNNLSTM), and Hyperband-Autoencoder (HB-AE), along with various oversampling methods, to rapidly and accurately detect FDIA. The following questions should be addressed:
1. The images attached to the article are blurry. The characters in some images are too small to be recognized. Please modify the images.
2. Table 1 appears somewhat cluttered and could be appropriately edited and streamlined. The strengths and weaknesses in the table should be separated into distinct sections to avoid presenting only the advantages or the disadvantages. Table 3 can directly use a confusion matrix figure, which is more intuitive than a table.
3. Section 3.1 appears to overlap with content from Chapter 1 and could be revised accordingly.
4. The equation font style in lines 246 and 255 differs slightly from that in lines 261, 265, and 267. Please ensure consistency in formatting throughout.
5. In line 339, “The architecture of the tuned CNN-BLSTM is shown in Figure 4(b).” Figure 4 is not labeled with (a) and (b).
6. The three pseudo-code segments could be reorganized into subsections 5.1, 5.2, and 5.3. Additionally, formatting inconsistencies exist across these segments; some instances of identical elements are bolded while others are not, and the presentation of conclusions varies between methods. Please ensure consistent formatting throughout. In this section, I suggest emphasizing a clearer explanation of the innovative aspects of the work.
7. In lines 492 and 496, formulas (13) and (14) should end with multiplying by 100% instead of 100.
8. In line 508, it says, “Performance analysis of unbalanced data without tuned layers of deep learning architectures such as ANN, CNN-LSTM, and Autoencoder is shown in Figure 5.”Why not compare ANN with RS-ANN? I think this better suits the research purpose of the paper.
9. The content described in Section 6.5 does not align with the information presented in Table 5. Additionally, Chapter 6 contains ten subsections (6.1 to 6.10), which appear redundant. Consider consolidating or streamlining these sections to improve clarity and conciseness.
10. Each chapter requires a heading, but Chapters 3 and 6 of this paper lack them.
11. The overall structure and level of detail in the entire paper require adjustments.
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.