Parallel Particle Swarm Optimization Using Apache Beam

Round 1

Reviewer 1 Report

Paper information-1568673 “Parallel Particle Swarm Optimization Using Apache Beam”

Comments

This study focuses on parallel particle swarm optimization using apache beam. I think the paper fits well the scope of the journal and addresses an important subject. However, a number of revisions are required before the paper can be considered for publication. There are some weak points that have to be strengthened. Below please find more specific comments:

*The abstract seems to be adequate. The only suggestion I have here is to explain a bit more “MapReduce”.

*Keywords: I suggest adding one or two most relevant keywords.

*Please fix the issue with cross-references. In many places, the manuscript shows “Error! Reference source not found”.

*Page 1: “An effective algorithm for solving optimization problems is particle swarm optimization (PSO)” – before discussing PSO specifically the authors to create a general discussion regarding the importance of artificial intelligence methods (e.g., heuristics, metaheuristics) for challenging decision problems. There are many different domains where artificial intelligence methods have been applied as solution approaches, such as online learning, scheduling, multi-objective optimization, transportation, medicine, data classification, and others. The authors should create a discussion that highlights the effectiveness of artificial intelligence methods in the aforementioned domains. This discussion should be supported by the relevant references, including the following:

• An online-learning-based evolutionary many-objective algorithm. Information Sciences 2020, 509, pp.1-21.
• Exact and heuristic solution algorithms for efficient emergency evacuation in areas with vulnerable populations. International Journal of Disaster Risk Reduction 2019, 39, p.101114.
• A proposal for distinguishing between bacterial and viral meningitis using genetic programming and decision trees. Soft Computing 2019, 23(22), pp.11775-11791.

Such a discussion will help improving the quality of the manuscript significantly. After discussing the use of artificial intelligence methods in different domains, it would be logical to focus on PSO specifically.

*Section 2: Description of the PSO seems kind of short. I suggest making it a bit more detailed.

*Perhaps sections 3 and 4 can be combined, since both of them are related to Apache Beam.

*Section 5: Please provide more references to justify the selection of input data that were used in the manuscript for computational experiments. This will be helpful to the future readers.

*The manuscript contains quite a lot of figures. Please double check to make sure that all figures are adequately described in the manuscript to prevent any confusion of future readers.

*The conclusions section should expand on limitations of this study and future research needs. I suggest listing the bullet points.

Comments for author File: Comments.docx

Author Response

I apologize for responding now because of the Chinese New Year visit to relatives and the redesign of running experiments. I have revised the manuscript according to your revisions, and the review comments response is attached.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper introduces (to the best of reviewer's knowledge) - a new parallelization approach to a PSO algorithm. First: the editorial quality is absolutely unacceptable: All links to citations are "Error! Reference source not found."  How it is possible to tolerate such a negligence without reaction form the editor?

Should I guess which works are cited?

Despite many shortcomings the work seems to be original, loosely related to Apache Spark parallelization of PSO at https://www.mdpi.com/2078-2489/12/12/530. 

Before deeper analysis of the paper, the links must be corrected, the "commercial type" text in lines 133-148 should be rewritten.

The drawback of the proposed approach is described as "the communication between particles is not sufficient (There is no guarantee that the neighbors of the current particle can receive the information sent by the current particle in time.), so the result is usually worse than MPSO."

The question arises: how much it could be worse? Because even if the execution time is lower -if the result is wrong - this can make the whole procedure useless. You should check and compare the accuracy of two methods 

-and-

"This can be improved by eliminating and updating the mechanism, so that particles that are not at the optimal value can be re-randomized into the 
feasible solution space, which is the direction of future work."

Why don't you extend the research and propose an efficient solution to this problem in this paper? The reader could think that maybe the whole approach is erroneous and can't be corrected in the proposed way or the proposed way offers no significant improvement over MRPSO in terms of the accuracy of results?

A practical application could be also useful , like in above cited paper.

Author Response

I apologize for the delay due to the Chinese New Year visit to relatives and redesign of running experiments. I have carefully checked and revised the manuscript according to your review comments, and the review response is described in the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors took seriously my previous comments and made the required revisions in the manuscript. The quality and presentation of the manuscript have been improved. Therefore, I recommend acceptance.

Author Response

I'm glad you took the time to review my revision and give specific comments. Isn't that how science keeps progressing? I'm glad you pointed out my shortcomings, and I will keep improving with your comments. Once again, thank you very much.

Reviewer 2 Report

The authors thoroughly improved the paper from the scientific point of view, including most of my comments. However the language and editorial quality is below expectations, many mistakes, missing spaces, etc.

Author Response

I'm sorry that my last revision still didn't satisfy you. I accepted all the changes in the revised draft and then rechecked the formatting issues such as spaces, as I am Chinese and may not be as fluent in English as a native speaker, but I have tried very hard to do so. I corrected all the similar errors that I could find. Thanks again for your comments, it will help me to do better.