A Link Prediction Algorithm Based on GAN
Round 1
Reviewer 1 Report
Abstract: Abstract needs to be improved following problem under study, research methods, results, conclusion and significance.
4 dataset used in this research Facebook, Wiki-Vote, Cooperative network CA-GrQc and cellular metabolic network Metabolic. Better to include Instrumental design and ethical considerations or approval etc.
More discussion required with the result.
Formatting and style of paper requires restructuring.
Extensive literature review from recent years is required as latest articles are from 2019 and some outdated references are used.
Referencing style needed to be checked and made according to journal format.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Dear Editors,
Dear Authors,
The reviewed study entitled: “A Link Prediction Algorithm Based on GAN Networks” represents interesting insight to the informatics. The reviewed study proposes new algorithm for link prediction within examined networks. The theoretical outline of the presented algorithm and presented experiments results seems to prove the algorithm’s computational efficacy as well as the ability to run on large-scale networks and to preserve the higher order structural features of network. It looks like the study does not offer any on-line platform or software devoted for the use of the developed algorithm by other researchers. The reviewed manuscript is correct in terms of substantive content, however; the study is difficult to read and understand for not experienced readers. Unfortunately, there is lack of supplementary data with example data processing step by step that could be additionally review. Moreover, there is no information how to use the developed algorithm by researcher in other study applications. It would be also nice to better explain for readers the terms and the equations presented in the manuscript as only a narrow group of specialists can understand it. The manuscript seems to be not corrected checked by native speaker and thus require many corrections (especially language) - some significant corrections have been done by myself. All questions and corrections are placed directly in the main text of manuscript (attached).
In conclusion, I recommend re-evaluating the manuscript for possible publication in the Electronics periodical after major revision with accordance to the remarks placed in the attached pdf file. Thank you for another interesting manuscript that I could review!
Comments for author File: Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
1. Introduction and related work is too short. Author need to include state of the art techniques as well.
2. How graph is layerd in structure.
3. Rest the paper is good but the idea is not novel.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Round 2
Reviewer 1 Report
I am happy with the extensive changes made by the authors. it can be published.
The response made by the authors "These four datasets are all open source datasets, so there is no need to worry about conflicts of interest and ethical considerations.". In my opinion it will be better that the article clearly states that to avoid the confusions.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Dear Editors,
Dear Authors,
The revised manuscript has been essentially improved and in my opinion is ready for publication in the Electronics periodical. The reviewed study characterizes by very good quality and the employed measurements against specified aims are rich and adequate. Language presentation is very well.
It is a very good job! Congratulations!
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
