A New Imbalanced Encrypted Traffic Classification Model Based on CBAM and Re-Weighted Loss Function

Round 1

Reviewer 1 Report

The paper clearly defines the problem of imbalance traffic data and the way to address it through DL methods.  The results and observations has been carried out on the publically available dataset and the proof is shown clearly for the solution to address the problem.  This paper in this state is appreciable and recommended.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors presented a novel method for classifying imbalanced encrypted traffic nonetheless, the paper requires minor correction.

The abbreviation CBAM-CEFL used in the title is unclear to prospective readers.

In figure 1, the Data Processing Module is a preprocessing module, and the three processes should typically occur in sequence order.

In Equation 11, describe the F1 Score metric, which might be understood as F1 minus score.

The authors should discuss the ResNet architectures employed and explain why similar architectures, such as VGG16/19, were not used...

Based on the base work, the proposed network framework resembles an AAE (Auto Encoders ) or GAN (Generative Adversarial Networks) architecture. More benchmarking on similar architectures and datasets is required to justify the suggested technique.

 

The introduction should be rewritten to avoid plagiarism from the paper “ET-BERT: A Contextualized Datagram Representation with Pre-training Transformers for Encrypted Traffic Classification” paper.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Dear Editor,

Here is my report on the paper applsci-1886897.

After reading the paper and reviewing it well, I see that the paper is well written and the topic is presented in an excellent manner. In order for this work to come out in its best form, I suggest to the authors some minor modifications:

1- Improving the introduction by presenting some recent papers in this field.

2- Review the punctuation marks in the entire paper.

3- Improving the conclusion section and showing the results they have reached.

4- I think that some modern references should be added in the introduction section related to the topic of the study and published in this journal. Add a new closely-related references:

*A numerical method for solving the nonlinear equations of Emden-Fowler models, Journal of Ocean Engineering and Science, 2022. doi.org/10.1016/j.joes.2022.04.019.

In the end, it was strongly recommended to accept the paper for publication.

Sincerely,

 

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx