Prediction of the Yield Strength of RC Columns Using a PSO-LSSVM Model

Round 1

Reviewer 1 Report

The submitted manuscript on “Prediction of the yield strength of RC columns using PSO- LSSVM model” has presented a good work. However, there are few shortcomings as listed below that need to be revised by the author(s) before any further recommendation –

1. What is the main question addressed by the research? This needs to be included in the Introduction section. There are so many objectives claimed; however, same should also be specifically drawn herein.  

2. The literature presented in introduction section need to be presented in the Tabular form as it is difficult to follow. It needs to be expanded further particularly other models used to predict yield strength of RC columns needs to be outlined clearly.

3. The discussion section 4.4 is too small. Literature support to this section needs to be inculcated with an overall improvement as well as division into 2-3 subsections.  

4. Add future recommendations for further investigation as well.

*Author(s) should highlight all the modifications carried out in the paper.

       

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The manuscript presents a new hybrid machine learning technique for predicting the yield strength and displacement of reinforced concrete (RC) columns. The proposed method is compared with other metaheuristic algorithms based on the experiment database selected in this study.

The paper needs a revision before reconsideration for publication in the journal Applied Sciences.
The following comments should be considered by the authors when revising the paper:
1)    The English language should be improved. The manuscript should be reviewed by a native English speaker with technical background to improve its quality.
2)    The introduction is a bit long and need to be shortened. Although a review of the scientific contributions from the literature is much appreciated, the authors should mostly restrict their attention to papers in which artificial intelligence tools are applied to structural engineering contexts (and not applied to other contexts in general). Indeed, this is not a review paper on the use of PSO, SI, SVM or enhanced variants. It is advised to move some text to specific subsections in the following parts of the manuscript, instead of reporting all these details in the introduction. This would improve the readability of the paper.
3)    The sentence at line 146 is incomplete. Please check and integrate it appropriately.
4)    With regard to the database used in Section 3, ref. [52] is a database dated 2004. In recent years, many other tests were conducted in the literature that could be added. The authors are referred to the following database:
Ghannoum W, Sivaramakrishnan B, Pujol S, Catlin A, Fernando S, Yoosuf N, Wang Y. NEES: ACI 369 Rectangular column database. 2015.
Publicly available at:
https://datacenterhub.org/resources/255
and updates at:
https://www.designsafe-ci.org/data/browser/public/designsafe.storage.published/PRJ-2526
Even if the authors do not wish to broaden the database, I think these additional tests should be at least quoted for the sake of completeness.
5)    The bibliography should be integrated by including hybrid models combining mechanical concepts with machine-learning-calibrated coefficients. The following recent paper should be quoted and commented in the field of shear strength capacity prediction of RC members (including RC columns and beams):
Quaranta, G., De Domenico, D., & Monti, G. (2022). Machine-learning-aided improvement of mechanics-based code-conforming shear capacity equation for RC elements with stirrups. Engineering Structures, 267, 114665.
6)    In terms of computational effort, the authors should provide more quantitative information on the comparison among the various algorithms examined in this comparative study (reported in Table 4).
7)    How do the authors think that the results of this study can be useful for other researchers and, above all, to practitioners involved in design tasks? Could the authors propose some easy-to-use regression formulae that, even in an approximate manner, can be based on the results obtained from the proposed PSO-LSSVM model? This would certainly increase the quality and scientific impact of this work.
8)    Do the authors intend to share their developed codes? Is there a public repository?

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Author(s) have done well. In my opinion, the paper is now ready for publication.  

Comments for author File: Comments.docx