Deep Learning Models Applied to Prediction of 5G Technology Adoption
Round 1
Reviewer 1 Report
In this manuscript, three models of deep learning were used to predict 5G technology adoption. Although the author tried to implement those models, the following comments should be considered.
--The abstract should be rewritten. It should contain answers to the following questions: What problem was studied and why is it important? What methods were used? What are the important results? What conclusions can be drawn from the results? What is the novelty of the work and where does it go beyond previous efforts in the literature?
-- The author indicated in the abstract that four models of deep learning were used, but they are three, not four. Also, you mentioned that “Deep Recurrent (DR)” is used as one of the deep learning methods, but in Section 3 you brought “Deep Reinforcement Learning” please clarify.
-- The abbreviations must appear at the very first place that the terminology is introduced and the way of introducing the terms must be consistent throughout the manuscript from abstract to conclusion. For example, (CNN) in Line 13 should be in Line 12 after Convolutional Neural Network.
-- The Introduction should make a compelling case for why the study is helpful along with a clear statement of its novelty or originality by providing relevant information and providing answers to basic questions such as: What is already known in the open literature? What is missing (i.e., research gaps)? What needs to be done, why, and how? Clear statements of the novelty of the work should also appear briefly in the Abstract and Conclusions sections.
-- The author should point out the major contributions of this paper by using 3 to 5 brief bullet points at the end of the Introduction section, right before the last paragraph.
-- The structure of arguments needs to be improved. At the end of the introduction part, you should have a section plan (for example section 2 discusses... and section 3 gives...).
--The Related Work Section may be elaborated to be more extensive to include a few recent works done by researchers who work on the traffic analysis defense domain because although 5G technology provides a robust security architecture for wireless network communications, it is insufficient to defeat traffic analysis (TA) attacks. Consequently, deploying robust, efficient TA defense solutions is essential. To improve this part the papers: A game-theoretically optimal defense paradigm against traffic analysis attacks using multipath routing and deception, and Quantifying mechanical properties of automotive steels with deep learning based computer vision algorithms should be discussed.
--Highlight all assumptions and limitations of your work.
--The author should provide other applications of the deep convolutional neural networks such as in Efficient deep neural networks for classification of Alzheimer’s disease and mild cognitive impairment from scalp EEG recordings, and Object pose estimation using mid-level visual representations.
-- Please pay attention to all punctuation marks.
-- What processor was used for the training process?
-- The Presentation of the Results and Discussion Section needs to be improved in the following aspects, (i) Insights should be presented regarding why the proposed approach performed much better than the existing methods. (ii) The experiments can be designed in a more elaborate way to cover all important aspects of the proposed approach.
-- What has been used in the current models to avoid the model overfitting problem?
-- The superiority of the proposed models should be addressed perfectly in comparison with similar established ones in the literature.
-- The relevant discussion and content of Figures and Tables are not enough.
--Conclusions: Please write some of these elements: open issues and probable solutions and/or applications, evaluation and implications of the work results or findings, and hypothesize a results explanation.
Author Response
I am indebted to you for drawing my attention to the error that I made. I am quite grateful that you read my work in its whole and provided feedback. It is quite significant to me, so thank you very much for bringing it to my attention.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
1. The abstract has presented the background, problem statement and previous studies' limitations. However, the authors failed to emphasize the findings, novelty, and future research directions. In addition to this, the abstract needs to be concise, as it exceeds more than 250 words.
2. The novelty and contribution of the study must be included in the introduction section. The organization of the study needs to include at the end of the introduction section.
3. A comparison of the previous works needs to be done in the literature review for a detailed analysis of the review for concluding the research gap.
4. The numerical findings of the study must be highlighted in the conclusion section. Need to provide a brief about the findings of the study for future research.
Author Response
I am indebted to you for drawing my attention to the error that I made. I am quite grateful that you read my work in its whole and provided feedback. It is quite significant to me, so thank you very much for bringing it to my attention.
Author Response File: Author Response.pdf
Reviewer 3 Report
This manuscript performed a comparative analysis of some selected deep learning models for the prediction of 5G adoption. While the authors provide good coverage of deep learning models, several aspects of the paper need to be improved.
#1. The contribution of the manuscript is clear and the analysis of the results is voluminous and difficult to comprehend. The authors should clearly show the contribution of this paper.
#2. The manuscript has to be thoroughly proofread for grammatical and typo errors as well as statement repetitions starting from the abstract.
e.g in the abstract
….This is due to the fact that the majority of the earlier research that has been done does not
14 use deep learning in connection with 5G adoption.
….The reason for this is based on the fact that the majority of the earlier research that has been done does not use deep learning in 18 connection with 5G adoption
…when 5G networks are adoptions are studied.
…cross-validation with thourouput
#3. There are two sections 3.1.2 and 3.2 in the manuscript.
#4. There are several Figure 5 and all the figures are not clear. The authors have to replace them with clear ones.
Author Response
I am indebted to you for drawing my attention to the error that I made. I am quite grateful that you read my work in its whole and provided feedback. It is quite significant to me, so thank you very much for bringing it to my attention.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The author addressed my comments and I have no further comments. My recommendation is to accept the paper at this stage.
Author Response
Reviewer’s 1 Comment
The author addressed my comments and I have no further comments. My recommendation is to accept the paper at this stage.
Reviewer’s 3 Response
I am very thankful that you have acknowledged that all of your corrections have been made and that you are pleased with the corrections that I have made, and I thank you very much for that. I am quite grateful to you.
I have also read the entire paper for spell check and adjusted the paper as you recommended.
Author Response File: Author Response.pdf
Reviewer 3 Report
One particular aspect is in Table 3. The caption "...thourouput" is showing there but now am satisfied with others.
Author Response
Reviewer’s 3 Comment
One particular aspect is in Table 3. The caption "...thourouput" is showing there but now am satisfied with others.
Reviewer’s 3 Response
I am quite grateful to you for making such a good insight, and I thank you very much for that.
The caption "...thourouput" is now removed in the Table 3
I have also read the entire paper for spell check and adjusted the paper as you recommended
Author Response File: Author Response.pdf
