Locally Specified CPT Soil Classification Based on Machine Learning Techniques

Round 1

Reviewer 1 Report

This is a relatively new article on the subject, focusing on the use of C4.5 models to improve soil classification in small areas.

1.     P193-194.The sentence “The following Figure 9 demonstrates the model trained by the combination of C#4 193 and T#4, which showed the best performance among the training results above, to com-194 pare with the Robertson’s diagram.” explains the comparison between Figure 1 and Figure 9, so for the convenience of readers' observation, it is suggested to put Figure 1 and Figure 9 together for the convenience of comparison.

2.     You should draw the flow chart of C4.5 participating in CPT Soil Classification for readers to observe.

3.     Double-check citations in the manuscript, e.g. P257, P253, etc.

4.     Pay attention to the resolution of the picture, some pictures in the text can not be seen clearly after enlargement. You should change the resolution of the picture in the text.

5.     Figure 2 is recommended to number the subgraphs, such as (a), (b), etc.

6.     Is the second figure in Figure 2 (from left to right) to express the distance between the sampling point and Hwajeon area 88 in Busan, South Korea? If so, it is suggested to place Hwajeon area 88 in Busan, South Korea at the origin of the coordinate? And then mark out the other distances. If it is not, it is recommended to modify it briefly.

7.     The tables in the manuscript should be used in a three-line table format, which is a more attractive and three-dimensional appearance.

8.     For Figure 4, there is no part about the performance of C#2 over other data sets in the article, so it is suggested to add this part in the section of Analysis Results.

9.     The expression of the model in the paper only shows the accuracy. And you are suggested to add the chart representation of the error of the model and other contents.

10.  If possible, you should add graphical abstract to increase the readability of the article.

 

Author Response

We appreciate the valuable and thoughtful comments from the reviewers. We tried to revise the manuscript following the remarks and prepare the answers for the comments.

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

The main question
Locally Specified CPT Soil Classification Based on Machine Learning Techniques Is.The introduction is described correctly, the article suggests a local approach and in my opinion, it can solve a problem because it is on a real scale. It is an article suitable for printing. The data has been processed correctly.
But the charts, especially charts 2, 3, and 6, need a clearer explanation. so as not to distract the reader from the main topic.
The results of the article have been processed correctly.
But the amount of content is very large and it is equal to one chapter of the book. It would have been better to avoid additional data.

In terms of written language, it needs minor changes.
In general, the article is suitable for publication with a few changes.

Author Response

We appreciate the valuable and thoughtful comments from the reviewers. We tried to revise the manuscript following the remarks and prepare the answers for the comments.

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Locally Specified CPT Soil Classification Based on Machine 2 Learning Techniques

Line 11: Define CPT in abstract as well:

Abstract is poorly written, if doesnt offer the much of work done and results attained, Modification required.

Introduction section: offers data mining and its techniques, it is not the introduction of the title you presented, so modify it.

Suggestion is to merge section 1 and section 2 as one.

Figure 1 can be removed, and its reference can be given for better understandings.

Table 4, datasets, from where these datasets are considered, is not clear, kindly mention the sources of datasets used here.

Is it publicly available or created by you, if you created this dataset then kindly provide their links.

Figures 5 to 8, need to be explained for the kind of accuracy they are offering with logical reasoning, why they are offering such accuracy.

Conclusion is not upto standards,

Paper is poor needs a lot of calibration as per standard SCI papers.

 

 

 

 

 

 

Author Response

We appreciate the valuable and thoughtful comments from the reviewers. We tried to revise the manuscript following the remarks and prepare the answers for the comments.

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Many changes are done.

Put some of the recent works in references as well.

After these changes can be considered.

Recognition of leaf disease using hybrid convolutional neural network by applying feature reduction

Weed density estimation in soya bean crop using deep convolutional neural networks in smart agriculture

 

 

 

Author Response

We appreciate the valuable and thoughtful comments from the reviewers. We tried to revise the manuscript following the remarks and prepare the answers for the comments.

Please see the attachment.

Author Response File: Author Response.pdf