Prediction of Swelling Index Using Advanced Machine Learning Techniques for Cohesive Soils
Round 1
Reviewer 1 Report
This is an interesting article and it is very well written. Thank you for all the details provided in the article. These details are the right dose; they are not too much, and sufficient enough to understand the methodology.
Author Response
We would like to express our sincere gratitude to Reviewer #1 for the time dedicated to the review and the constructive and positive remarks.
Reviewer 2 Report
You have suggested that several attempts have been made for estimating the vital swelling index parameter (and have listed two disadvantages: expensive and time-consuming). A justification is not provided. A thorough analysis may provide at least a table indicated this issue.
Author Response
We would like to express our sincere gratitude to Reviewer #2 for the time dedicated to the review and the constructive remarks, which allowed us to improve the quality of our article, deepening the research and potentially helping it reach a broader audience.
We agree with the reviewer #2 that the original manuscript did not pay sufficient attention to clarifying the disadvantages of the Oedometer tests and gathering the attempts published in the literature for estimating Cs. However, in the revised manuscript we have carefully revised the introduction by adding a satisfactory explanation supported by references. Now we have better explained the disadvantages of the Oedometer tests in the introduction (see page 1 and page 2 (lines 40-48) in the revised manuscript). Furthermore, as suggested by the reviewer, we have added a new table (Table 2, see Line 89) which recapitulates the attempts made in the literature for estimating the vital swelling index parameter using ANN method for supporting Table 1 which presents just the attempts to estimate Cs by simple correlations. The changed text was marked using a yellow background, in order to allow for easily spotting the changes.
Reviewer 3 Report
The manuscript is very interesting and important for geotechnical research. The methodology is set out clearly, but the article itself is presented in a very messy way. It is necessary to reformat the entire text, including tables and equations.
TABLE 1:
Column names are started with small letters. Equations are written in two rows. Near references could be written the authors of equations.
Rows 69, 74, 77:
It is not clear what the authors wanted to say with numbers in parentheses (example (2-8-1) on 69 row).
Rows 129, 131, 223, 230:
Equations go beyond the text or are unnecessarily written in two lines.
Rows 158, 182, 349...:
The titles of the subsections do not differ from the text.
Rows 421, 423, 426, 427, 429 ....:
very small symbols, the same symbols are written as regular and italic.
Row 427:
Written incorrectly FS-RF = 1,07.
Tables 2, 3, 4, 5, 9 and figures 2, 3 are wider than the text area.
Author Response
We would like to express our sincere gratitude to Reviewer #3 for the time dedicated to the review and the comprehensive, profound and constructive remarks, which allowed us to improve the quality of our article, deepening the research and potentially helping it reach a broader audience. In this regard, we approached every comment, hopefully, in a proper and exhaustive manner; there are some comments where we need more clarifications in order to address them. The text below presents in detail the way of addressing each comment; the references are to the final line numbers of the revised article. In addition, the changed text was marked using a yellow background, in order to allow for easily spotting the changes.
Author Response File: 
Author Response.doc
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
