Prediction of Grain Yield in Wheat by CHAID and MARS Algorithms Analyses

Round 1

Reviewer 1 Report

This manuscript (agronomy-2395416) aimed to determine the correlations among different wheat species' morphological characteristics and demonstrate the application of MARS and CHAID algorithms to a wheat data set. The algorithms identified agronomic qualities that positively influence grain production as indirect selection criteria and determined that the MARS model was the best model to predict grain yield.

Minor comments:

L39 and 43. What is “8 term”?

L43. What is “smallest RMSE”:

I suggest that the authors rewrite the abstract section on the main results and conclude with future perspectives in agriculture regarding the applicability and contribution of their work. This will provide a better understanding of the study's potential impact and significance for the field.

Please change the keywords to different ones than those in the title and put the new keywords in alphabetical order.

Please consider rewriting the first paragraph to be more direct and concise.

L107. Italic in scientific name; check all manuscript;

Change the table to a location map.

Table 2. What is the Directorate of Meteorology? Please provide more information.

Table 3. Why only 0-20 cm?

Change 2.4 section. “Method” not adequate.

What type of correlation was used in Table 3? Pearson, Spearman? Was it standardized or p values reported?

For Figure 3, it is unclear from the manuscript what the dendrograms represent in relation to the numbers. It would be helpful to provide a legend or additional information to explain the dendrograms and their relationship to the genotypes/species. In general, it is important to provide complete and clear information for each figure and table in the manuscript, so that readers can understand the data presented without having to refer back to other parts of the manuscript. X-axis “common” not correct, change for “dot”.

Figure 4. I do not understand what the authors are trying to convey with the figure. Moreover, its quality is low, and I do not see its relevance to the manuscript. Please consider removing it or creating a supplementary file.

L324. In substance?

Check old references;

 

Major corrections in grammar and spelling are needed. Additionally, many paragraphs need to be shortened for better clarity.

Author Response

Responses to Comments of Editor

 

General Response:

 

Dear editor; According to the valuable suggestions of the commentators, the comments of Reviewer 1 on the manuscript are highlighted in yellow. Thank you for giving us the chance to review our manuscript. In the table below, we tried to respond to the suggestions and comments of all the referees in the best way possible.

 

Reviewer 1

This manuscript (agronomy-2395416) aimed to determine the correlations among different wheat species' morphological characteristics and demonstrate the application of MARS and CHAID algorithms to a wheat data set. The algorithms identified agronomic qualities that positively influence grain production as indirect selection criteria and determined that the MARS model was the best model to predict grain yield.

Comment	Response
1.       L39 and 43. What is “8 term”?	The expression "8 term" is a result of MARS analysis, as presented in table6. The MARS model presented an 8-term formula for estimating grain yield. The abstract section of the manuscript was rewritten and incomprehensible sentences were corrected (L39).
2.       L43. What is “smallest RMSE”:	Root-mean-square error (RMSE). While MARS and CHAID algorithms offer the best prediction model, they are based on the lowest RMSE value. The abstract section of the manuscript was rewritten and incomprehensible sentences were corrected (L43).
Minor comments: I suggest that the authors rewrite the abstract section on the main results and conclude with future perspectives in agriculture regarding the applicability and contribution of their work. This will provide a better understanding of the study's potential impact and significance for the field.	The abstract section in the manuscript has been rewritten considering the reviewer's suggestion.
Please change the keywords to different ones than those in the title and put the new keywords in alphabetical order.	The keywords section in the manuscript has been rewritten considering the reviewer's suggestion.
Please consider rewriting the first paragraph to be more direct and concise.	The first paragraph in the manuscript has been rewritten considering the reviewer's suggestion.
L107. Italic in scientific name; check all manuscript;	All scientific names in the manuscript have been checked and corrected.
Change the table to a location map.	A location map has been added under Table 1.
Table 2. What is the Directorate of Meteorology? Please provide more information. Table 3. Why only 0-20 cm?	The climate data for the Igdir region in Table 2 are sourced by the Turkish State Meteorological Service. In response to the reviewer's suggestion, the manuscript was revised to include additional information for Table 2.   Table 3 has been visually reorganized
Change 2.4 section. “Method” not adequate.	Changed the 2.4 section title to Measured characteristics of dependent and independent variables. Added 2.5 section (CHAID and MARS analysis).
What type of correlation was used in Table 3? Pearson, Spearman? Was it standardized or p values reported?	Pearson was used in Table 3. The relevant expression has been added to the appropriate place in the material and method section. p values were reported below the figure. New written expressions are as follows; Pearson’s pairwise correlations between wheat traits were computed by R Studio software (Allaire, 2012) by the ’corrplot’ package (Wei et al., 2017).
For Figure 3, it is unclear from the manuscript what the dendrograms represent in relation to the numbers. It would be helpful to provide a legend or additional information to explain the dendrograms and their relationship to the genotypes/species. In general, it is important to provide complete and clear information for each figure and table in the manuscript, so that readers can understand the data presented without having to refer back to other parts of the manuscript. X-axis “common” not correct, change for “dot”.	Figure 3 has been rearranged considering the review suggestion.
Figure 4. I do not understand what the authors are trying to convey with the figure. Moreover, its quality is low, and I do not see its relevance to the manuscript. Please consider removing it or creating a supplementary file.	Figure 4 displays the tree diagram tree diagram produced by the CHAID algorithm. Tree diagrams are used to visualize the data obtained as a result of the analysis. The numerical data visualized on the tree diagrams represent the descriptive data of the model. The sentence "Figure 4 displays the tree diagram produced by the CHAID algorithm" has been added to the manuscript for reference to the tree diagram. Considering the review recommendation, Figure 4 is given as supplementary material (Supplementary Figure S1).
L324. In substance?	“In substance” has been changed to “correlation result between variables”
Check old references;	Old references checked considering the review suggestion, and replaced by current literature.

 

 

Author Response File: Author Response.docx

Reviewer 2 Report

The study adopted multivariate adaptive regression splines (MARS) algorithm and chi-square automatic interaction detector (CHAID) for estimation yield of Wheat. Please consider the following issue in the revised manuscript.

In Line 227, it is better to have the Discussion section separately to compare the experiment results with those of other studies. In addition, authors' opinion or intuition can be provided in that section.

 

In Line 336, the Conlusion section needs to be extended with limitations of the study. In addition, a summary of Expriments involving numerical results and Discussion can be addressed in Conclusion.

Author Response

Responses to Comments of Editor

 

General Response:

 

Dear editor; According to the valuable suggestions of the commentators, the comments of Reviewer 2 are highlighted in blue. Thank you for giving us the chance to review our manuscript. In the table below, we tried to respond to the suggestions and comments of all the referees in the best way possible.

 

Reviewer 2

In Line 227, it is better to have the Discussion section separately to compare the experiment results with those of other studies. In addition, authors' opinion or intuition can be provided in that section.	Review suggestion was taken into consideration and "Results" and "Discussion" sections were separated. The "Discussion" sections have been expanded as much as possible.
In Line 336, the Conclusion section needs to be extended with limitations of the study. In addition, a summary of Experiments involving numerical results and Discussion can be addressed in Conclusion.	"Conclusions" section has been edited considering the review suggestion

 

 

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

I would like to thank the authors for responding to my comments (agronomy-2395416). However, the captions for the tables and figures are still not appropriate. They require many modifications to truly contextualize the reader as to what they refer to. No reader should be obliged to return to the text or interpret what the figure or table is showing, without clear guidance. Therefore, please rewrite all of them with as much information as possible.

For example, "Table 4. Descriptive Plant Measurement Statistics." It is necessary to provide context and define each variable. I "do not know" what GNS, GY... are, similar for table 5.

Topic 5 conclusions.

Minor corrections.

Author Response

Dear Reviewer;

According to the valuable suggestions of the commentators, the comments of you on the manuscript are highlighted in yellow .Thank you for giving us the chance to review our manuscript. 

Author Response File: Author Response.docx