Integrating Blockchain and Deep Learning for Enhanced Mobile VPN Forensics: A Comprehensive Framework
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsAbstract is missing. It can be seen from the submission form, but the manuscript is missing abstract which is good to have.
Introduction: Please simplify first sentence of third paragraph that starts as "On the flip side as...". It is very hard to follow.
Paragraph 6, which is considered to the problem statement of this manuscript focusses on is very vague. Please be specific with details. Name those statistical analysis, machine learning techniques you are referring to.
2.1 Digital Forensics and Blockchain
Blockchain technology has been a game changer across almost every industry, say, health care, vehicle tracking etc. However, that statement alone doesn't make it a strong case to self-cite several blockchain related papers. Likewise, about Android malware analysis.
Implementation: Reduce font size of code or wrap text according to page size. Several lines are cut in the current paper, and it is hard to follow.
Results: Results are more analytical when relative parameters are considered. For example, the same data sets before and after application of your approach is what I'm looking fir as a reader/reviewer. Proof that your approach is better when compared to existing ones is missing.
Proof/ base of these figures is missing. Analysis of non-VPN vs VPN traffic is not the point under discussion in this manuscript.
Conclusion: By the end of this paper, focus has slowly shifted from security to classifications of VPN networks.
References: I see a lot of citations which are not relevant to this study. These must be deleted, especially the citations from author Toqeer Ali (total 6 citations self-cited some of which are not appropriate). Please explain how citations 6, 20, 21 are applicable to the current topic.
Author Response
Thanks for your valuable review on our paper. We did our best to incorporate your requests and recommendations.
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe paper provides a method of VPN traffic classification in a forensic setting using graph neural networks and convolutional neural networks. The authors state that their solution offers a complete solution for VPN traffic classification, enabling forensic investigators to distinguish between VPN and non-VPN traffic pertaining to a particular application. However, for me, this statement is not fully justified by necessary statistical tests. Additionally, the authors employed blockchain but do not clearly describe why they need it and its characteristics. The system design was also not fully justified. Why do they use Hyperledger Fabric instead of, for example, Exonum? Another issue is the lack of an introduction, poor formatting, and a focus only on the Saudi Arabia case. Additionally, extensive self-citation is detected, with references 5-7 and 19-22. There are only 23 references in total. At the same time some new papers and approaches are missing, for instance paper https://ieeexplore.ieee.org/document/9681767. Another issue is the source code, which takes up almost two pages. I recommend that the authors upload the source code to GitHub and provide a link to it. Taking these issues into account, I cannot recommend the acceptance of this paper.
Comments on the Quality of English LanguagePlease proofread the paper one more time and correct any issues with articles.
Author Response
Thanks for your valuable comments, we tried our best to incorporate the changes requested. The review response file (point by point) also includes the reviewer 1 responses too so you have an idea what we did with overall paper.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsPleased with the new version. I'm glad the author addressed all the previous comments.
Author Response
Thanks for accepted our replies to the reviews.
Reviewer 2 Report
Comments and Suggestions for AuthorsThe author did a good job incorporating my comments; however, I cannot recommend paper acceptance at the current stage due to the following drawbacks:
- 1. The authors completely ignored my comment regarding comparison with the paper https://ieeexplore.ieee.org/document/9681767.
- 2. A link to the source code on GitHub was not provided. It is good practice to include the link to the source code in the published paper to ensure easy access to it. This fact will definitely result in an increase in the value of the paper and its citations. Additionally, in my opinion, the paper would benefit from a link to the ISCXVPN2016 dataset.
- 3. The authors did not add any justification for choosing Hyperledger Fabric within the text. Please add it.
Author Response
Thanks for your valuable comments. please see attached file, the comments are addressed.
Author Response File: Author Response.pdf