FLIP: A New Approach for Easing the Use of Federated Learning

Round 1

Reviewer 1 Report (New Reviewer)

It would be interesting to make a comparison of the proposed solution with the solutions presented in related works.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 2 Report (New Reviewer)

The paper overall is quite good. The paper present FLIP as novel approach.

However, I suggest to authors to do these changes

1- No need for use case

2- Did you design figure 2, if no cite it

3- Figure 3 title is too long shorten it

4- conclusion is too long move some of it to the main text

.......

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report (New Reviewer)

This article presents FLIP (Federated Learning Interactive Platform), which has been developed as a comprehensive, easy-to-use fully functional web-based FL network management platform. However, some essential concerns need to be addressed, which are listed below:

(1) I cannot find clearly the motivation and technical contribution of the work. Either the mechanism and experimental results are not clear.

(2) Could you provide more details on the implemented federate learning algorithm?

(3) Can you provide a comparison table comparing the proposed method with related state-of-the-art works?

(4) The format and organization of the article are suggested to be improved. 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report (New Reviewer)

All comments are done

Reviewer 3 Report (New Reviewer)

All my concerns have been addressed in the current version.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

-        The paper introduces the Federated Learning Interactive Platform (FLIP) that could provide ease in using federated datasets in a real environment while preserving some legal and ethical requirements. In saying that, there are several areas that could be improved by the authors.

-        Firstly, the contributions of the papers includes identifying the different roles for the proposed platform and the ease of user interaction. Perhaps the contribution of the paper could be more specific on the proposed solution, especially on how the platform could automate the activities needed for FL.   

-        Secondly, with the proposed FL platform, additional information should be added on how the authors derived the design for the model. Even though the proposed model might be a new idea but it would seem the approaches and design is derived from previous literature.  

 

-        Thirdly, for the validation of the platform, how was the approach done while being aligned with he concept of FL? What data was used by the platform in predicting the new cases of Covid-19 and what was the file size that was used for the aggregated data?  Further information on this could improve the paper as well.  Overall, it is a good paper but it could be improved with further information. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Interesting data and discussion

Author Response

Dear reviewer, thank you very much for your positive feedback. We are very pleased that it was of interest to you.

Reviewer 3 Report

·       What are the real difficulties that justify author's work, i.e., what are the most important challenges authors want to handle for pixel-wise mechanical damage detection of fresh-eaten corn? Why is it so difficult? I suggest to state this information clear in the introduction in order to give a better understand of the work.

·       Please add a Related Work section in your paper to display some published work related to your study. This can help readers to understand your work. An in-depth overview of the work of classification should be further given. Therefore, the reviewer suggests discussing the advances by citing some references, e.g.,

·       I recommend authors present some papers to show briefly different methods and their feasibility to apply to the studied subject. This can make reader to extend his/her view about the importance of your job. need more convinced literature reviews to indicate clearly the state-of-the-art development.

1.     New modified controlled bat algorithm for numerical optimization problem

2.     Detection of fake news text classification on COVID-19 using deep learning approaches

3.     A fine-tuned BERT-based transfer learning approach for text classification

 

·       The authors just compared their method with different classifiers. More comparisons with state-of-the-art classification methods are needed.

·       There exist some grammar issues in the paper including the choices of words an the sentence structure, which disturbs the readability. A native English speaker is strongly recommended for this task to polish the language and correct the grammar errors

·       I think the objectives are clearly stated in the manuscript, the authors need just to highlight clearly their contribution in the abstract.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

·        The authors have taken considerable efforts to respond to the comments I had submitted earlier. However, I do apologize in saying that the paper still lacks the validation for the results which is very critical to check the contribution of this work compared to the current state of arts.

·        Related work section is very short. I suggest adding some more comparison of existing methods.In this study, it is necessary to make a comparison with the existing methods in the literature. If necessary, additional experimental studies should be added and the results should be shared in order to make comparisons.
Conclusion needs reconsideration. It needs to highlight more the research main contribution, briefly answering questions
'how ?' with some untrivial indications and improvement percentages to keep with the reader. Also, the conclusion needs to present some more ideas of open research and future work for researchers to build upon for further advancements.

·        It would be useful to review the studies I recommend in your future studies. I recommend authors present some papers to show briefly different methods and their feasibility to apply to the studied subject. This can make reader to extend his/her view about the importance of your job. need more convinced literature reviews to indicate clearly the state-of-the-art development. Consider the following studies in the reference section:

1.            Comparative Analysis of Low Discrepancy Sequence-Based Initialization Approaches Using Population-Based Algorithms for Solving the Global Optimization Problems

2.            A Systematic Literature Review on Particle Swarm Optimization Techniques for Medical Diseases Detection

·        The authors need to follow the journal format for citing the references. Reference 84, 86, and 87 are incomplete.

·        There are grammatical mistakes in the manuscript, the authors need to proofread it from native English speaker. Authors should use clearer and concise vocabulary to express their ideas and discussions.

·        The authors need to add extend discussion of the results. Discussion should include comparisons with other existing approaches in literature. The contributions seem to be reasonable to justify the publication for this journal.