Shear Sonic Prediction Based on DELM Optimized by Improved Sparrow Search Algorithm
Round 1
Reviewer 1 Report
My comments:
1. Abstract - "it can be inferred that the method adopted in this research could be a productive" - I'm not sure if the authors wrote the abstract before writing the paper. Put some quantified results here, which would be more useful than this vague sentence.
2. Several of the acronyms are not explained or misleading. e.g. DTS. I'm not from this field of oil/gas exploration, so I'm not sure what to make of this.
3. The literature review is sparse and the paper only has a dozen references. Please include a more updated literatur review instead of citing papers from 2004 and 2011.
4. Please proofread your paper. There are numerous typing errors, random capitalizations, missing spaces, etc. e.g. bette instead of better
5. SSA - What is SD? Do not use multiple symbols for physical quantities. Subscript or superscript.
6. Please provide a table or graphic with all neural network parameters.
7. Figure 7 is a bit misleading. The difference in the performance of various algorithms has not been quantified even though it can be easily done. It's about 0.7%. Why should the reader think this is a significant improvement?
8. Figure 9 is used to assert the superiority of ISSA. But based on what? The authors just assert that it can be visually observed. Quantify the improvement. Otherwise the plots all look the same to me.
9. Conclusions - no numbers are given.
10. The authors suddenly bring up shortcomings in the method, something about stratum feature. The last couple of lines of the paper is not the place to introduce new information.
Author Response
Response to Reviewer 1 Comments
Point 1: Abstract - "it can be inferred that the method adopted in this research could be a productive" - I'm not sure if the authors wrote the abstract before writing the paper. Put some quantified results here, which would be more useful than this vague sentence.
Response 1: The quantitative prediction results have been added according to your requirements, and the corresponding words have been modified. Please see the modified abstract for details.
Point 2: Several of the acronyms are not explained or misleading. e.g. DTS. I'm not from this field of oil/gas exploration, so I'm not sure what to make of this.
Response 2: The sonic log is also called the compressional wave time difference log, and DTC is short for compressional wave time difference log. The shear sonic log is also called shear wave time difference log, and DTS is short for shear wave time difference log.
Point 3: The literature review is sparse and the paper only has a dozen references. Please include a more updated literature review instead of citing papers from 2004 and 2011.
Response 3: The references have been updated according to your request. Please refer to the resubmitted article for details.
Point 4: Please proofread your paper. There are numerous typing errors, random capitalizations, missing spaces, etc. e.g. bette instead of better.
Response 4: I rechecked the whole article and revised the corresponding mistakes. Please review the resubmitted article for details.
Point 5: SSA - What is SD? Do not use multiple symbols for physical quantities. Subscript or superscript.
Response 5: I have made modifications according to your requirements, please see the resubmitted article for details.
Point 5: You say that it should be pointed out that although the BO-HKELM model proposed in this research has achieved excellent results, the input feature set does not consider the factors such as stratum structure, so it cannot completely reflect the real underground information. I want to learn why did not you include geological factors in this work.
Response 5: Due to the requirement of confidentiality, it is difficult for us to get geological information of specific blocks. We will communicate with related oil fields in the future, hoping to obtain some relevant information. But we can't do that right now.
Point 6: Please provide a table or graphic with all neural network parameters.
Response 6: According to your requirements, the corresponding table has been added to this paper. Please refer to the resubmitted article for details.
Point 7: Figure 7 is a bit misleading. The difference in the performance of various algorithms has not been quantified even though it can be easily done. It's about 0.7%. Why should the reader think this is a significant improvement?
Response 7: As shown in the figure 7, the ISSA converges more quickly than other algorithms. Furthermore, the final error resulted from ISSA is lower than those obtained from other algorithms. All these can show the superiority of ISSA algorithm. The correlation between the other logs and the missing DTS curves is high, so there is less room for optimization using the algorithm..
Point 8: Figure 9 is used to assert the superiority of ISSA. But based on what? The authors just assert that it can be visually observed. Quantify the improvement. Otherwise the plots all look the same to me.
Response 8: I have added the corresponding R-square to each figure to help readers discover the improvement in prediction results through ISSA optimization.
Point 9: Conclusions - no numbers are given.
Response 9: In accordance with your request, quantitative prediction results have been added to the conclusion of this paper. For details, please check the resubmitted article.
Point 10: The authors suddenly bring up shortcomings in the method, something about stratum feature. The last couple of lines of the paper is not the place to introduce new information.
Response 10: I deleted the shortcomings of this article and made a prospect for the next step of the research plan. Please refer to the resubmitted paper for details.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear Editor:
Considering the manuscript with the ID: applsci-1831700, entitled "Shear Sonic Prediction Based on DELM Optimized by Improved Sparrow Search Algorithm", herewith I would like to submit my comments. I can say that the manuscript is well organized and well referenced, the introduction and the methodology are well explained. Nevertheless, some more discussions still need to be added to improve the comprehensiveness and demonstration of the method and some needs to be removed, revised or edited. Some minor issues need to be addressed or corrected to improve the general quality and readability of this article. So based on my comments that comes in the following, I proposed the paper to be subjected for moderate revision.
Best Regards
General Comments:
Is the paper new, technically correct, and relevant?
Yes, the paper is new and technically sounds. Results support the methodology, but needed to be more cleared by the author in case of experiment assumptions.
Is the paper well organized?
The paper is properly organized, good literature review, suitable motivation and clear explanation on results are positive points to that.
Is the abstract concise?
Yes, but I think it might need to be rephrased after revision to add some comments about any artifacts or negative points in the method, if exist.
Is the introduction motivating?
Yes, Introduction section is motivating.
Are the methodology, results, and conclusions completely developed?
No, they need to be modified and developed according to the technical comments.
Are there language, mathematics, reference, or style errors? There is no mathematical, reference or style error.
Technical Comments:
Are the codes available for this research? As I found, there is no code available for this study, e. g. in Github. If the authors could make the codes available, the manuscript could be much better evaluated, not only for reviewers, but also for possible readers. When it is not possible to upload the code for public access, such as in Github, could they be provided for reviewer for better assessment of the study?
The authors should explain what limitations did they find out about the proposed method.
How did you evaluate the final result? How did you consider to finally selection a methodology for data with more inconsistency between logs or complexity in relationship between logs?
The authors should explain what limitations did they find out about the proposed method or selected model in Sparrow Search Algorithm.
The abstract focusses mainly on the general problem and ignores the other items such as the methodology, good introduction, results and conclusion.
The introduction section is a nice one. It is architected very beautifully, while written fully academic and comprehend. I assume that any change in the introduction section is not necessary, but one of the important tasks after publishing a study is to increase its chance to be seen by the most possible number of researchers, so I would like to give two recommendations. First, to get your published study in the list of searched for papers based on keywords, I propose to increase variety of your keywords. In my viewpoint, they do not cover the whole topic of the study and are not widely searched words. I propose to add at least the keyword of the machine Learning concepts. Second, one of the methods in the publisher’s website that brings a publication on to the researchers, is based on the similar publications that they have read before. So, the more you cite similar publication, the more the chance that the search engine in the publisher website propose your paper to the researcher. Besides of that, it will also complete your introduction section. As another advantage, it rises new ideas to the researchers by combining various methods, or resolving drawback of one seen paper by reading the similar one, or extending the methodology to a fully automatic one. So, based on these points, I would like to ask to cite to the following similar publication in the manuscript. The first proposed publication that could be cited in the introduction is: Shahbazi, A., Soleimani Monfared, M., Thiruchelvam, V., Ka Fei, T., Babasafari, A.A., (2020). Integration of knowledge-based seismic inversion and sedimentological investigations for heterogeneous reservoir. Journal of Asian Earth Sciences, 202, 104541, in using artificial intelligence methods in prediction of porosity and permeability and the second one is: Soleimani, M., and Jodeiri, B., (2015), 3D static reservoir modeling by geostatistical techniques used for reservoir characterization and data integration. Environmental Earth Science, 74, 1403–1414. You can cite this publication considering near the same topic in reservoir characterization.
Best Regard
Comments for author File: 
Comments.pdf
Author Response
Response to Reviewer 2 Comments
Point 1: Are the codes available for this research? As I found, there is no code available for this study, e. g. in Github. If the authors could make the codes available, the manuscript could be much better evaluated, not only for reviewers, but also for possible readers. When it is not possible to upload the code for public access, such as in Github, could they be provided for reviewer for better assessment of the study?
Response: The relevant code is part of the relevant fund project. We have applied for disclosure, but the feedback we got was that it could not be released to the public for the time being.
Point 2: The authors should explain what limitations did they find out about the proposed method.
Response: In this method, the firefly disturbance strategy is used to update the position of all individuals in the population, which makes the algorithm search more fully in the solution space, effectively avoids falling into the local optimum, and then improves the convergence speed and accuracy of the algorithm. However, the diversity of population locations did not increase, which would also affect the accuracy of the method to a certain extent.
Point 3: How did you evaluate the final result? How did you consider to finally selection a methodology for data with more inconsistency between logs or complexity in relationship between logs?
Response: The significance of this paper is that compared with other conventional machine learning methods, the missing data prediction method proposed in this paper is faster and more accurate. This method can be used in any block, as long as the lithology and the geological background is similar.
Point 4: The authors should explain what limitations did they find out about the proposed method or selected model in Sparrow Search Algorithm.
Response: Sparrow search algorithm (SSA), as a novel swarm intelligence optimization algorithm, has been proved to have better search performance. However, due to the decrease of search ability and population diversity in the middle and late iteration of SSA in some cases, the algorithm has some shortcomings, such as slow convergence speed, low solution accuracy and easy to fall into local optimal solution.
Point 5: The abstract focusses mainly on the general problem and ignores the other items such as the methodology, good introduction, results and conclusion.
Response: I have modified the relevant parts according to your requirements. Please see the resubmitted article for specific modifications.
Point 6: The introduction section is a nice one. It is architected very beautifully, while written fully academic and comprehend. I assume that any change in the introduction section is not necessary, but one of the important tasks after publishing a study is to increase its chance to be seen by the most possible number of researchers, so I would like to give two recommendations. First, to get your published study in the list of searched for papers based on keywords, I propose to increase variety of your keywords. In my viewpoint, they do not cover the whole topic of the study and are not widely searched words. I propose to add at least the keyword of the machine Learning concepts. Second, one of the methods in the publisher’s website that brings a publication on to the researchers, is based on the similar publications that they have read before. So, the more you cite similar publication, the more the chance that the search engine in the publisher website propose your paper to the researcher. Besides of that, it will also complete your introduction section. As another advantage, it rises new ideas to the researchers by combining various methods, or resolving drawback of one seen paper by reading the similar one, or extending the methodology to a fully automatic one. So, based on these points, I would like to ask to cite to the following similar publication in the manuscript. The first proposed publication that could be cited in the introduction is: Shahbazi, A., Soleimani Monfared, M., Thiruchelvam, V., Ka Fei, T., Babasafari, A.A., (2020). Integration of knowledge-based seismic inversion and sedimentological investigations for heterogeneous reservoir. Journal of Asian Earth Sciences, 202, 104541, in using artificial intelligence methods in prediction of porosity and permeability and the second one is: Soleimani, M., and Jodeiri, B., (2015), 3D static reservoir modeling by geostatistical techniques used for reservoir characterization and data integration. Environmental Earth Science, 74, 1403–1414. You can cite this publication considering near the same topic in reservoir characterization.
Response: I have quoted the similar publication in the manuscript according to your requirements. Please see the resubmitted article for specific modifications.
Author Response File: Author Response.docx
Reviewer 3 Report
The subject investigated in this paper is in the scope of the journal. However, both readability and presentation of the paper should be checked and improved.
The original contribution of the paper is not clearly stated. The contributions of the article must be emphasized in terms of originality, significance, and performance metrics in the abstract and introduction.
Recently-proposed metaheuristics from high impact factor journal (see https://www.scimagojr.com/) should be cited like from IEEE transactions, Springer and Elsevier in the introduction or in a related work section.
Examples:
https://ieeexplore.ieee.org/document/8672851
https://ieeexplore.ieee.org/document/6819057
https://ieeexplore.ieee.org/document/8721125
https://www.sciencedirect.com/science/article/abs/pii/S0957417420305224
https://www.sciencedirect.com/science/article/abs/pii/S0965997820301484
Fig. 2,5,7,8, 9, among others are with low resolution. Please verify it.
Results in Table 3 could be based on nested cross-validation. performance measures and significance nonparametric tests. Authors could perform statistical tests (e.g. Friedman test + posthoc Nemenyi test) to compare algorithms/models and discuss the results in the paper.
State-of-art metaheuristics such as CMAES (http://cma.gforge.inria.fr/) and coyote algorithm (https://github.com/jkpir/COA) could be considered in the comparative study with the proposed optimizer.
Author Response
Response to Reviewer 3 Comments
Point 1: The subject investigated in this paper is in the scope of the journal. However, both readability and presentation of the paper should be checked and improved.
Response: I rechecked the whole article and improved the readability and presentation. Please review the resubmitted article for details.
Point 2: The original contribution of the paper is not clearly stated. The contributions of the article must be emphasized in terms of originality, significance, and performance metrics in the abstract and introduction.
Response: I have revised the relevant parts according to your requirements, please see the resubmitted article for specific changes.
Point 3: Recently-proposed metaheuristics from high impact factor journal should be cited like from IEEE transactions, Springer and Elsevier in the introduction or in a related work section.
(see https://www.scimagojr.com/)
Examples:
https://ieeexplore.ieee.org/document/8672851
https://ieeexplore.ieee.org/document/6819057
https://ieeexplore.ieee.org/document/8721125
https://www.sciencedirect.com/science/article/abs/pii/S0957417420305224 https://www.sciencedirect.com/science/article/abs/pii/S0965997820301484
Response: I have quoted the relevant literature you requested. Please refer to the resubmitted article for details.
Point 4: Fig. 2,5,7,8, 9, among others are with low resolution. Please verify it.
Response: I have improved the resolution of the relevant pictures, please check in the resubmitted article.
Point 5: Results in Table 3 could be based on nested cross-validation. performance measures and significance nonparametric tests. Authors could perform statistical tests (e.g. Friedman test + posthoc Nemenyi test) to compare algorithms/models and discuss the results in the paper.
Response: The relevant table shows the relevant results of cross-validation, which I will re-emphasize in the article. I will make use of your proposed method for algorithm comparison in the next research work. Thank you for making this idea. We will work on the statistical tests in the future, and you will see this study in our upcoming work.
Point 6: State-of-art metaheuristics such as CMAES (http://cma.gforge.inria.fr/) and coyote algorithm (https://github.com/jkpir/COA) could be considered in the comparative study with the proposed optimizer.
Response: Thank you for your suggestion, this is what we need to work on in the future, we are also working on this, you will see this study in our future work.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The authors have made several changes and improvements to their manuscript. They have responded to each comment in a satisfactory manner. However, I am still concerned about a couple of things.
-They claim to have provided a table with the neural network architecture, but both versions (old and revised) contain exactly three tables. I'm not sure what to make of this. Where is the new table?
-Overall significance of the study, which seems very niche and project-like, may not appeal to a wide readership, perhaps a more niche journal should be approached.
Due to the overall quality not being up to the standards of this journal (reference are still lacking, quality of plots, results), in addition to the issues raised above, I recommend rejecting this article. If the editor wants to override my decision, I defer to them, but please look into the missing table.
Author Response
Response to Reviewer 1 Comments
Point 1: They claim to have provided a table with the neural network architecture, but both versions (old and revised) contain exactly three tables. I'm not sure what to make of this. Where is the new table?.
Response 1: I have described the network structure adopted in this paper in Section 3, see the resubmitted article for details.
Author Response File: Author Response.docx
Reviewer 2 Report
Dear Editor:
Considering the manuscript with the manuscript ID: applsci-1831700, entitled " Shear Sonic Prediction Based on DELM Optimized by Improved Sparrow Search Algorithm", herewith I would like to submit my comments. I can say that the author(s) have addressed all my comments and my concerns about the uncertainty analysis, evaluation of the model and validation of results which are resolved and well explained in the revised manuscript. So, based on the improvement of the manuscript in the revised version, I can say that it is appropriate for publication. So, my final decision is that the manuscript can be accepted upon the final decision of the editor.
Best Regards
Author Response
Thank you for your appreciation！
Reviewer 3 Report
All acronyms must be defined in the abstract including RMSE, MAPE, and R-square.
Sugestion:
which provides a promised method for DTS estimation.
->
which provides a promised method for DTS estimation.
The remainder of this paper is organized as follows. Section 2 presents .... Section 3 ...
Recent references about ELM and DELM could be mentioned. Some examples are:
https://www.sciencedirect.com/science/article/abs/pii/S0925231218304260
https://www.sciencedirect.com/science/article/abs/pii/S0893608020302598
https://www.sciencedirect.com/science/article/abs/pii/S0306261919307974
State-of-art metaheuristics such as CMAES (http://cma.gforge.inria.fr/), Mirjalili approaches (https://seyedalimirjalili.com/projects) and coyote algorithm (https://github.com/jkpir/COA) could be considered in the comparative study.
A full statistical analysis of the optimizers comparison must be presented based on significance nonparametric tests.
Authors could perform statistical tests (e.g. Friedman test + posthoc Nemenyi test) to compare algorithms and discuss the results in the paper.
See examples in the following papers:
Recent trends in the use of statistical tests for comparing swarm and evolutionary computing algorithms: Practical guidelines and a critical review
Swarm and Evolutionary Computation Volume 54 May 2020Article 100665
J. Carrasco, S. García, M. M. Rueda, S. Das, F. Herrera
https://www.sciencedirect.com/science/article/pii/S2210650219302639
Analyzing convergence performance of evolutionary algorithms: A statistical approach
Information Sciences Volume 28924 December 2014 Pages 41-58
Joaquín Derrac, Salvador García, Sheldon Hui, Ponnuthurai Nagaratnam Suganthan, Francisco Herrera
https://www.sciencedirect.com/science/article/pii/S0020025514006276
Author Response
Response to Reviewer 3 Comments
Point 1:
All acronyms must be defined in the abstract including RMSE, MAPE, and R-square.
Sugestion:
which provides a promised method for DTS estimation.
->
which provides a promised method for DTS estimation.
Response: All acronyms have be defined, and the definition of RMSE, MAPE, and R-square can be seen in section 4.4. Please see the resubmitted article for details.
Point 2:
The remainder of this paper is organized as follows. Section 2 presents .... Section 3 ...
Response: These have been added in the section1 (the introduction). Please see the resubmitted article for details.
Point 3:
Recent references about ELM and DELM could be mentioned. Some examples are:
https://www.sciencedirect.com/science/article/abs/pii/S0925231218304260
https://www.sciencedirect.com/science/article/abs/pii/S0893608020302598
https://www.sciencedirect.com/science/article/abs/pii/S0306261919307974
Response: I have quoted the relevant literature you requested. Please refer to the resubmitted article for details.
Point 4:
State-of-art metaheuristics such as CMAES (http://cma.gforge.inria.fr/), Mirjalili approaches (https://seyedalimirjalili.com/projects) and coyote algorithm (https://github.com/jkpir/COA) could be considered in the comparative study.
A full statistical analysis of the optimizers comparison must be presented based on significance nonparametric tests.
Authors could perform statistical tests (e.g. Friedman test + posthoc Nemenyi test) to compare algorithms and discuss the results in the paper.
See examples in the following papers:
Recent trends in the use of statistical tests for comparing swarm and evolutionary computing algorithms: Practical guidelines and a critical review.
Swarm and Evolutionary Computation Volume 54 May 2020Article 100665
- Carrasco, S. García, M. M. Rueda, S. Das, F. Herrera
https://www.sciencedirect.com/science/article/pii/S2210650219302639
Analyzing convergence performance of evolutionary algorithms: A statistical approach
Information Sciences Volume 28924 December 2014 Pages 41-58
Joaquín Derrac, Salvador García, Sheldon Hui, Ponnuthurai Nagaratnam Suganthan, Francisco Herrera
https://www.sciencedirect.com/science/article/pii/S0020025514006276
Response: Thank you for this valuable feedback. Relevant references have been added, please see the resubmitted article for details. In this research, the data of Well M are used to establish the model by cross validation, and the established model is applied to Well N. It has a good prediction effect on both test set and validation set. Due to the limitation of experimental conditions and insufficient computing resources, there is no way to do more tests at present.
Author Response File: Author Response.docx
