Analysis of Preprocessing Techniques for Missing Data in the Prediction of Sunflower Yield in Response to the Effects of Climate Change

Round 1

Reviewer 1 Report

Line 61, A background session should contain the rationale, the key problem statement, and a brief overview of research questions that are addressed in the rest of the paper. Some unnecessary past studies, such as disease detection (see ln.69) and a proper time for yielding (even if a part of this is close to your study see in 75.) take place in this section. The closer studies to this paper should be represented, or just remove since too much information exists. 

Line 177, create a new paragraph underlying the overall limitation of the study, at the simplest level, the weaknesses of the study, based on factors that are often outside of your

control as the researcher. 

Table 1. The order of the papers is confusing. The authors should sort the data in ascending order according to their published years. 

Ref.27, The given link presents the category, data item and geographic level of sunflower yielding. Indicate what has been selected for this study as a footnote.  

Line 140, Be clear on why you shrink the data. 

Eq.1 The AVGWeek is expressed in the text, but it is not available in the equation.  On the other hand, let's say the area planted  a few times in a specific land through (wi – wi-1). How can you take this into account in the equation? 

Ref. 28 Express the credibility of the Data source (Daymet Online Source) that is used for this study.  Is it complete or any missing part? Was it used by another study before?

Section 3.2 should also include the name and version of the software program,  the processing time for the calculation and the spec of the computer that the author used (if any intense programming exists).

Ln. 224, nonsense, remove the expression “From a mathematical point of view,”

Section 3.2.2. is not mentioned clearly in Figure 2. Revise

Figure 5 should be discussed more, underlying the behaviour of LOf, SVM-OC, IF and EE against the applied contamination rates

Figure 6 is not mentioned in the text. Insert an explanation.

Line 309, convince the author 3000 iteration is enough. Was it tested through the algorithms/formulas as calculation of Power Spectra or numerical result analysis? If done, insert. 

Line 312, See my comment above, insert the spec in 3.2

Line 364, Express the striking correlation for the data presented through Figure A1. 

Indicate an explanation for Figure 7. Figure 7 has two circles over HGBR(LOF) and HGBR(MCD). I am not sure what they are. Explain them. 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

There are following observations that should be incorporated before consideration for publication as given below:

1. The choice of preprocessing techniques for missing data seems arbitrary and lacks justification. The authors should provide a clear rationale for selecting the specific techniques used in the study.

2. The experimental design lacks rigor and fails to establish a fair comparison between the preprocessing techniques. The authors should consider using a randomized controlled trial or a cross-validation approach to ensure unbiased and reliable results.

3. The evaluation metrics used to assess the performance of the models are limited and do not provide a comprehensive understanding of the predictive accuracy. Additional metrics such as precision, recall, F1-score, or area under the ROC curve should be included to provide a more robust evaluation.

4. The sample size used in the study appears to be small, which raises concerns about the generalizability of the findings. The authors should consider increasing the sample size or conducting the study on a larger dataset to improve the statistical power and reliability of the results.

5. The discussion of the results is superficial and lacks a critical analysis of the limitations and implications of the findings. The authors should provide a more in-depth interpretation of the results, addressing any inconsistencies or unexpected outcomes.

6. The paper lacks a clear theoretical framework or conceptual basis for the choice of preprocessing techniques. It would be helpful for the authors to provide a literature review or theoretical background to support their selection of the techniques used.

7. The conclusion is vague and does not provide a clear summary of the findings. The authors should explicitly state the implications of their results and highlight any recommendations for future research or practical applications.

8. The paper could benefit from a more thorough discussion of the limitations and potential sources of bias in the study. This would help readers understand the reliability and validity of the findings and provide insights for future research.

9. The paper lacks novelty and originality. The authors should clearly articulate the unique contributions of their study and how it adds to the existing body of literature on predicting crop yield.

Overall, the paper needs substantial revisions and improvements in terms of methodology, analysis, and interpretation of results. Addressing these concerns would significantly enhance the quality and contribution of the study.

Proof reading required

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Journal “Applied Sciences” - 2470136 manuscript examines An Empirical Analysis of Preprocessing Techniques for Missing Data in Predicting Sunflower Crop Yield. This manuscript describes new and carefully confirmed findings and experimental procedures which are given in sufficient detail. The length of a full paper satisfies the minimum required to describe and interpret the work. The manuscript is very well written, with relevant information that adequately addresses the given topic.

The style and language are mainly in keeping with academic standards, providing clearly expressed information in a logical order and easy to follow.

The manuscript is prepared according to journal writing instructions and needs a minor revision. The following comments were raised:

New Title: “Analysis of preprocessing techniques for missing data in the prediction of sunflower yield in response to the effects of climate change”.

Line 6: Please change “seeding date” to “sowing date” in the document.

Line 134: Please change “seeding” to “sowing”.

Line 152: Please be more explicit about “using daymetpy library”.

Line 160: Is a good number of researched years? 41 or 42? The 41 resulting features are the following: Year (1980 to 2021).

The conclusion should be adjusted based on the main findings of the study and provide guidelines for further research in the references, please complete and correct some of them. References - There should be a uniform style. See instructions to authors.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf