Investigation of the EWT–PSO–SVM Model for Runoff Forecasting in the Karst Area

Round 1

Reviewer 1 Report

The similarity rate of the study is 40%. It has a rate too high to be published in this journal. It also has an 8% similarity to the other authors' study. Under these conditions, the similarity rate of the manuscript should be decreased, and the manuscript should be resubmitted.

 

Report for ApplSci-2333010
The topic of the manuscript looks good. The layout and presentation of the manuscript needs crucial
improvements. Moreover, I recommend the following corrections to improve the manuscript:
1- The authors should have used a grammar check editor for some grammatical errors and misprints.
2- In Introduction Section, the major contributions of the manuscript should be detailed clearly.
3- In Introduction Section, the organization of this study should be provided.
4- Why the authors compare proposed method with at least one of Fuzzy SVM, Twin SVM, Fuzzy Twin
SVM, and Intuitionistic Fuzzy Twin SVM?
5- Why the authors compare proposed method with at least one of ABC-SVM, ACO-SVM, BOA-SVM,
and GA-SVM?
6- Conclusion section should be highlighted by mentioning advantages, disadvantages, and limitations
of the study. The future works should be extended also.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

This reviewer suggests the following points to improve the paper quality:

1.       Proofread the entire manuscript to rectify several existing typos and grammatical errors.

2.       Limit the acronyms in Abstract and Conclusion section (unless repeated within the section). Acronyms (EWT-PSO-SVM) should be defined/expanded at it’s first appearance.

3.       Try to include the nomenclature of all the symbols used in the work, at the beginning for better readability. Also list the design values of each system parameters in appendix.

4.       Try to redraft the Introduction section, with background, challenges, literature review, scopes, motivation, contributions, and organization of paper. Highlight the novelties/major contribution of the work prior to organization pf paper in brief (preferably in 3-bulleted points). Also try to expand the literature review including some recent works (of last 3-years) in the similar field.

5.       Try to maintain the workflow of the paper, especially during transition between sections and subsections.

6.       Try to emphasize more on the problem statement and objective of the work.  The validation of the proposed method should be supported by comparative analysis with the contemporary methods.

7.       Redraw all the figures with enhanced resolution and include XY grids to result plots for clarity in presentation.

8.       Try to quote all the equations in related texts with proper citation (if adopted from published work)

9.       Results should be supported with more comparative analyses for different scenarios/case studies to support the claim.

10.   Redraft the Conclusion with numerical evidence to support your claim. Also include at least one future scope to it.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Please answer the following questions:

Introduction:

1. What is the significance of accurate and reliable runoff forecasting in the field of hydrology, and why is it important for optimal allocation of water resources and flood control and disaster reduction?

 

2. What are some advantages and disadvantages of process-driven and data-driven models for runoff forecasting, and how do they differ in their approach to analyzing the runoff process?

 

3. What are some commonly used data-driven models in runoff forecasting, and what are some strategies that can be used to improve the predictive ability of these models, particularly in the context of capturing periodicity and regularity in the nonlinear and non-stationary characteristics of the runoff series?

 

Discussion:

1. How does the "decomposition-forecasting-reconstruction" approach improve the accuracy of model forecasting, and what is the role of EWT in this approach?

 

2. What is the role of the PSO algorithm in optimizing the parameters of the SVM model, and how does this optimization improve the generalization ability of the model?

 

3. What are some previous studies that have shown the effectiveness of the EWT-PSO-SVM model in forecasting, and how do the results of this study compare to those studies?

 

4. What are some potential reasons for the relatively unsatisfactory results of the developed EWT-PSO-SVM model in terms of the MAPE index, particularly in the context of using runoff data from karst basins?

 

5. What are some potential directions for future research in improving the forecasting performance of different hybrid models for accurate runoff forecasting in karst basins?

 

6. What is the EWT-PSO-SVM model, and how does it differ from previous hybrid models used to predict runoff in karst basins?

 

7. What is the role of EWT in improving the accuracy of runoff forecasting, and why is it a feasible method for use in this field?

 

8. What is the MR of the EWT-PSO-SVM model, and how does it compare to the other two models (SVM model and PSO-SVM model) in terms of forecasting accuracy?

 

9. How does the developed model perform in the forecasting of large flow events, and how can these results be useful in water resource allocation?

 

10. What is the MR of the EWT-PSO-SVM (M) model, and how does it perform under the monthly structure?

 

11. What are some potential areas for further research in improving the accuracy of runoff forecasting in karst basins, and what are some other artificial intelligence or machine learning methods that could be used in conjunction with EWT?

 

Conclusion

1. What is the purpose of the hybrid EWT-PSO-SVM model proposed in this study, and how does it aim to improve the accuracy of runoff forecasting in karst basins?

 

2. How does the EWT-PSO-SVM model work, and what are the steps involved in its "decomposition-forecasting-reconstruction" approach?

 

3. What are the main findings of this study in terms of the superiority of the EWT-PSO-SVM model compared to the single SVM model under different data structures, and what does this suggest about the effectiveness of data decomposition and parameter optimization strategies in improving the accuracy of SVM models for predicting nonlinear and non-stationary runoff series in karst basins?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Dear Author(s),

I have attached the review report. Please examine it.

Best regards, 

Comments for author File: Comments.pdf

Author Response

Dear reviewer(s):

Thank you very much for your guidance and we have responded based on your report. Please see the attachment for details.

Best regards,

Author Response File: Author Response.docx

Round 3

Reviewer 1 Report

Dear Authors,

 

In the revised manuscript, all the suggestions have been adopted and the major corrections have been made. Therefore, my evaluation is that the revised version of this manuscript can be published in this journal.

 

 

Best regards,