Tillage-Depth Verification Based on Machine Learning Algorithms
Round 1
Reviewer 1 Report
1. The title is also more relevant to the manuscript.
2. For data collection, four algorithms ( Kmeans,DBSCAN, GMM and curve fitting) were used, but the statistical comparison was between RANSCA and DBSCAN, where DBSCAN is a new algorithm.
3. I couldn't find any good and updated literature about these algorithms ( Kmeans, DBSCAN, GMM, and curve fitting), so please provide the updated literature.
4. Please make the problem more clear by breaking it down into points.
5. Because this is a research article, it would be better to provide a block diagram and flow chart of the proposed methods.
6. What is the difference between table 1 and table 2 as simulated soil samples and other field soil samples.
7. It would be better to provide curve representations of these algorithms as well.
8. References need to be updated.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
Authors have put significant effort to design and analyze Data analysis methods based on a machine learning algorithm that discriminates tills. However, the following suggestions could be included to improve the quality of this article.
1. Introduction section should focus on the following points such as introduction to the research area, motivations, objectives, and contribution. It needs to be improved.
2. Literature review section is also not clearly identified and written. The readability of the paper reduces due to a lack of explanation. The problem of your work is not clearly identified. Limitations of this method are also emphasized and how the application domain of your model should be justified by the requirements of real-world problems.
3. There is very limited in novelty of the work. It must be clearly specified in terms of "text and model". The proposed section is not clearly written. It should clearly specify the methodology and its working principles in detail.
4. More explanation is required in methodology rather than basic derivations. Some more standard datasets may be considered and the performance measure should be considered. 5. Results should be validated using some statistical techniques.
6. Different parameters that may be considered during the experiment, need to be clearly explained.
7. Result analysis should also be compared with some existing work to improve the quality of the article. All the methods must be tested through some standard datasets and results must be compared.
8. In the result analysis section, please specify, why in some situations accuracy is very low. Please specify
9. Some other performance measures such as precession, recall, and f-measure need to be computed and results must be presented.
10. In the conclusion section, the limitations and future scope of the method need to be clearly addressed.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors' article addresses the problem of determining the depth of the soil in the tilled layer by exploring several machine learning algorithms. The experimental work is well-designed, and the results are promising.
Some improvements needed
-The related work is intertwined with the introduction, and it's minimal.
Plus, the article's contribution to state-of-the-art needs to be clarified.
-The abstract and introduction need revision. Some grammatical typos need attention.
-The title structure is unconventional, but it's a matter of writing style.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 4 Report
The Paper is good, I have some suggestions given below
1. Expand the DBSCAN and GMM either in the abstract or in the introduction.
2. Can you try other machine learning methods to increase the efficiency of the model?
3. Why are only K means, DBSCAN, and GMM used from machine learning algorithms why not other methods? It will be beneficial if you give proper justification for choosing only the above three methods.
4. Make section 3,4 and explain the working of RANSCA
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 2 Report
It can be accepted for publication.