Wheat Soil-Borne Mosaic Virus Disease Detection: A Perspective of Agricultural Decision-Making via Spectral Clustering and Multi-Indicator Feedback

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The article explores a new intelligent decision model that combines AI with decision science to assess wheat soil-borne mosaic virus severity across diverse agricultural plots. This approach treats planting sites as virtual decision-makers and uses intuitionistic fuzzy numbers to capture the uncertainty in diagnostic outcomes. By integrating a Bayesian-GCN for trust propagation and enhanced spectral clustering for subgroup formation, the model effectively addresses ecological heterogeneity and spatial variability in symptom expression. Validation using the Walla Walla Valley dataset shows that this model outperforms traditional models in consensus efficiency and decision robustness, offering valuable insights for targeted disease management.In general, I think this paper is well suited for the issue named Advances in Artificial Intelligence for Plant Research and offers a new insight for the combination of plant modeling with AI.

 

To strengthen the manuscript, the authors should clarify the rationale for the parameter ranges used in the sensitivity analyses. For instance, in Section 6.2.1, while the results demonstrate stability within the range of α=0.38-0.9, a brief explanation of why these specific bounds were chosen would enhance the transparency of the methodology. Additionally, the introduction should more explicitly differentiate the virtual DM concept from conventional expert-centric LSGDM approaches, as this distinction is central to the model’s novelty. Regarding the presentation of tables and figures, Table 2 requires unified notation for planting sites, and the contradiction between the CRP time reported in Table 8 and the claim that removing trust propagation significantly alters rankings should be addressed. Furthermore, the axis labels in Figure 9 should clearly define the principal components shown in the t-SNE projection.

 

The authors should also briefly address the scalability of the model, particularly how computational demands, such as the 55 feedback iterations, might affect real-time deployment in larger agricultural networks. Including a sentence on computational complexity or hardware requirements would provide useful context for practical implementation. Finally, the authors should explicitly link the final rankings in Table 6 to actionable interventions, explaining how these rankings should prioritize resource allocation. These refinements will enhance the clarity and applicability of the manuscript, ensuring that its contributions remain well-supported by empirical validation.

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

- The authors should make it clear for what is the proposed model mentioned in the Abstract.

- The authors should add more small diseased pictures in the Figure 1 and annotate the Figure 1 with text of healthy areas and text of diseased areas.

- The authors should add more pictures for SBMV leaves and healthy leaves in the Figure 2, and annotate the Figure 2 with text of SBMV leaves and text of healthy leaves.

- The authors should merge above revised Figure 1 and revised Figure 2 into one Figure.

- The authors should revise the title of the Figure 4 or blocks in the Figure 4, since the title said the LSGDM is a framework while the Figure said the LSGDM is a block.

- The authors should check whether the Figure 4 and the Figure 5 are same, since their titles are very similar.

- The authors should merge the Figure 3 and the Figure 4 into one Figure. Please consider also whether should merge further with the Figure 5.

- The authors should reduce the number of so much steps. The authors should better use flow chart or pseudocode algorithm to represent some of these steps.

- The authors should check whether the Table 3 is a Table, and whether the Table 6 is a Table.

- The authors should not only use color but use shadow to distinguish different values in the Figure 11.

- The authors should increase the size of markers in the Figure 10.

- The authors should revise the title of the manuscript by reducing the length of the title and replacing the current part with more specific words that may be too broad in scope.

Comments on the Quality of English Language

Could be improved.

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

T

The manuscript try to describe possible application of AI in the forecasting of one virus plant disease. A partly annotated version is provided, however the manuscript must be drastically improved and moved to an appropriate journal such as Pathogens or similar considering that the content related to plants is minor.

The author must eliminate all the personalization and reorganize the manuscript in an appropriate structure as an experimental manuscript. Now it is written only in narrative form and quite difficult to follow and understand. Only data related to the subject must be reported and discussed avoiding unnecessary information about basic bioinformatic processes. More relevance must be done to the disease studied, avoiding assessing that the system proposed can be used for other pathogens since this is a very specific disease. More knowledge about this disease must be added to clarify the content of the application. The use of two years data is not sufficient to feed a model for the plant disease studied.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

No major comments for the English but see enclosed annotated version of the manuscript

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors did a good job. I have several minor revision comments, as follows.

- The authors should reorganize the Figure 1. The title of the Figure 1 said there are three on top right but it seems there are two on the top right. Besides, the number of scenes or images of health and WSBM is not equal. I would suggest the authors display both health images and WSBM images by same categories of long distance, medium distance, and short distance. I would suggest the authors display 1 image, 2 images and 4 images for long distance, medium distance, and short distance respectively for both health and WSBM. Each categories should use a small rectangle without borders and with a relatively deep background color, and same categories use same color. Besides, I would suggest the authors use two rows three columns for the overall layout. The first row is for the health images, and the second row is for the WSBM images. The first column is for the long distance. The second column is for medium distance, and the third column is for short distance. Texts of long distance, medium distance, short distance, health and WSBM should be involved and placed at suitable location. After that, two large rectangle without borders and with two relatively shallow but different background color should be adopted, and one is for health and the other is for WSBM.

- It is better to use three-line table for Table 5 and Table 7.

- It is better to use two lines to display text in each cell of Figure 6. For example, the original text of 0.375 could be divided into 0.3 for the first line and 75 for the second line. After above division, size of those small texts should be slightly enlarged.

- The authors should adjust the layout of Figure 2. The overall layout should be similar to a rectangle. Namely, the authors should arrange the elements on the right more loosely, or add more elements, or move the position of some elements, etc.

- The authors should better use a vector figure for Figure 7 since texts in the figure may be not very clear.

Comments on the Quality of English Language

Fine but could be polished.

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The revised manuscript take into account just a small number of required revisions. This is an almost complete rewriting, and the topic distribution are still very unclear. The manuscript is quite confusing and does not allow to evaluate its coherence with the aims described. Inconsistencies are present in the acronym of the disease studied and the pathological part is not quite well developed and used only a little to produce the results. The manuscript as it is just phantasy-plant pathology. A partially annotated version is provided to highlight some of the aspects to be improved.

Comments for author File: Comments.pdf

Comments on the Quality of English Language

To be improved to clarify the matter described.

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx

Round 3

Reviewer 3 Report

Comments and Suggestions for Authors

The revised version still needs some minor revisions that are marked in the enclosed annotated version.

Comments for author File: Comments.pdf

Author Response

Please refer to the attached file.

Author Response File: Author Response.docx