Content of review 1, reviewed on May 03, 2020
Comments on abstract, title, references
Abstract
The abstract is adequate in giving a glimpse of the background information about the issue under discussion and its importance.
Research gaps are not stated.
The aim of work is given clearly along with a lengthy, unnecessary details about the employed methodology.
Results are not provided, only a brief summary of conclusion is indicated.
Title
- The Suggested title is put accurately, it sounds fairly interesting as it refers to some of the key features of the work.
References
References do not follow any known referencing style, but are alphabetically sorted.
Not very current, many references are more than 15 years ago and more (to the year of publication 2009)
Most of the references are relevant to the subject.
Some major references were absent from referencing at that time (up to 2009), such as:
1- De Bondt, Michiel C. (November 2004). "NP-completeness of Master Mind and Minesweeper". Radboud University Nijmegen. 2- Greenwell, D. L. "Mastermind." J. Recr. Math. 30, 191-192, 1999-2000. 3- Andrew M. Singley, (2005),“Heuristic Solution Methods for the 1-Dimensional And 2- Dimensional Mastermind Problem”. MSc. Thesis, UNIVERSITY OF FLORIDA.
Comments on introduction/background
The introduction gives a very detailed description of the problem at hand, enforced by an example.
No related work was provided.
The authors stated the complexity of the problem being NP-Complete.
A research gap was roughly defined as: "however, there was no additional selection on the combination played other than the fact that it was consistent with the responses given so far."
The research purpose was clearly outlined and justified with clear objectives.
Comments on methodology
Subject selection is declared in a clear sequence.
The previous related methods are described along the explanation in a discontinuous manner, they should have been gathered in single section of related work.
No details or definition about evolutionary algorithms are given.
The fitness function should have been introduced and discussed in a separate section before the details of the approach.
Variables were invoked and measured suitably.
The discussion of the methods shows that an internal validation has been accomplished, however external validity is not confirmed. The study is, nevertheless, reliable.
There are enough details to reproduce the results and replicate the study, knowing that the study itself has some replications of other previous work.
Comments on data and results
Sample size has been fixed before the experiments.
Results are better placed in a separate section.
Table captions should be placed above the table not under it.
Title of table in table caption is missing for all tables in the article.
Table captions should not contain details of the carried test, these details ought to be explained in a separate paragraph.
The text in the results of the study dose add to the data.
The presented results are not sufficient and ends with a research gap statement: “How this proportion grows with search space size is still an open question”
Comments on discussion and conclusions
A contradiction exists between the abstract and the conclusion as the abstract states that: “this paper proves that by the incorporation of local entropy into the fitness function of the evolutionary algorithm it becomes a better player than a random one”. In the conclusion: "However, how this is incorporated within the evolutionary algorithm remains to be seen, and will be one of our future lines of work.”
The conclusion dose not answer the aims of the study, much is still suggested for future work.
The study has a number of limitations that are listed to be dealt with in future study.
Overall
The study design was not very appropriate in answering the aim stated in the study
This study focused on combining local entropy within the fitness function for the used evolutionary algorithm
The major flaws of this article were the missing related work section, the lack of sufficient info about the evolutionary method used, and the gained results.
The article lacks consistency between the abstract and the conclusion
Overall statement or summary of the article and its findings in your own words
1- The study presented a research gap defined as:” however, there was no additional selection on the combination played other than the fact that it was consistent with the responses given so far”, which establishes a wide range of investigations to be studied.
2- The NP-Complete problem was well defined.
3- The authors commenced on investigating various heuristics to improve the power of one of the evolutionary algorithms known as the Estimation of distribution algorithm (EDA) for guiding the search of the optimum from heuristic candidate solutions.
4- A reasonable testing process was carried out but the results were not sufficient and ended up with a research gap statement: “How this proportion grows with search space size is still an open question”
5- There are enough details to reproduce the results and replicate the study, knowing that the study itself contains some replication of other previous work.
6- The study has a number of limitations that are listed to be dealt with in future study.
7- The conclusion dose not answer the aims of the study, much is still suggested for future work.
Overall strengths of the article and what impact it might have in your field
1- The authors suggested a fairly interesting title for the study by the inclusion of some key features of the work.
2- The research purpose was clearly outlined and justified with clear objectives.
3- Subject selection is declared in a clear sequence.
4- Variables were invoked and measured suitably.
5- Sample size has been fixed before the experiments
6- The discussion of the methods shows that an internal validation has been accomplished, however external validity is not confirmed. The study is, nevertheless, reliable.
7- Most of the references are relevant to the subject.
Specific comments on weaknesses of the article and what could be done to improve it
Major points in the article which needs clarification, refinement, reanalysis, rewrites and/or additional information and suggestions for what could be done to improve the article.
1- The abstract can be enhanced by adding a knowledge research gap along with some insight of the gained results.
2- Authors are encouraged to provide a separate section for the related work of the study.
3- An explanation of Evolutionary Algorithms together with the mechanisms (reproduction, mutation, recombination, and selection) can have a great impact on clarifying the suggested method.
4- It is preferable that references follow a known referencing style, with the support of more current studies in the field.
5- Some major references can be added to reinforce the clarity of investigation, such as:
a. Greenwell, D. L. (2000), "Mastermind." J. Recr. Math. 30, 191-192.
b. Andrew M. Singley, (2005),“Heuristic Solution Methods for the 1-Dimensional And 2- Dimensional
Mastermind Problem”. MSc. Thesis, UNIVERSITY OF FLORIDA.
c. De Bondt, Michiel C. (2004). "NP-completeness of Master Mind and Minesweeper". Radboud
University Nijmegen.
6- There seems to be a contradiction between the abstract and the conclusion, the abstract states that “this paper proves that by the incorporation of local entropy into the fitness function of the evolutionary algorithm it becomes a better player than a random one”. And in the conclusion: “However, how this is incorporated within the evolutionary algorithm remains to be seen, and will be one of our future lines of work.”
Minor points like figures/tables not being mentioned in the text, a missing reference, typos, and other inconsistencies. 1- Some Equations lack numbering.
2- Table captions should be placed above the table not under it.
3- Title of table is missing for all tables in the article.
4- Table captions should not contain details of the carried test, these details ought to be explained in a separate paragraph.
5- The fitness function should have been introduced and discussed in a separate section before the details of the approach.
6- Results are better placed in a separate section.
Source
© 2020 the Reviewer.
References
Philip, R. T., J., M. J. 2010. Adapting Heuristic Mastermind Strategies to Evolutionary Algorithms. Studies in Computational Intelligence.