Data-Driven Soil Analysis and Evaluation for Smart Farming Using Machine Learning Approaches
Round 1
Reviewer 1 Report
Please see the attachment.
Comments for author File: 
Comments.pdf
good
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
The review of the manuscript „Data-driven soil analysis and evaluation for smart farming using machine learning approaches.
The topic of the manuscript is very interesting. The quality of the data and analyses is high; therefore, the manuscript is worth publishing.
I have comments regarding the discussion section that misses comparison with other studies. Are the results similar? Or are the findings different?
In the manuscript, there are 17 figures and 15 tables. It is a lot. Please limit this. Some figures and tables are not cited in the text.
In addition, the quality of some figures (3, 4, 7, 13, 14, 15, 16, 17) should be improved.
Author Response
1) I have comments regarding the discussion section that misses comparison with other studies. Are the results similar? Or are the findings different?
According to your suggestion, we added one paragraph about study comparison in the Discussion Section as following:
‘’8.2. Creative soil analysis platform confusion
The traditional way to collect soil data is field observation through crop report or sensors, which is time-consuming and high cost. The modern way is to measure passive land surface microwave emission and radar backscatter, which is more robust and reliable. Most soil analysis systems combine those data sources with geospatial data to visualize and monitor soil status.
Our nutrient analysis approach uses the result from the crop identification module and irrigation module, providing an efficient way to utilize physical and biological soil data with cost efficiency. Our platform is a creative system that combines multiple soil analysis steps into one work, benefiting small-scale farmers specifically. In the future, we will deploy customized research results based on local drone and sensors data.’’
2) In the manuscript, there are 17 figures and 15 tables. It is a lot. Please limit this. Some figures and tables are not cited in the text.
Thank you for the suggestion. We ran through all the figures and tables again, and cross-referenced the missing parts. And for the quantity of tables, we deleted some unessential tables in the Data Collection part.
3) In addition, the quality of some figures (3, 4, 7, 13, 14, 15, 16, 17) should be improved.
According your suggestion, we’ve re-edited those figures.
