Use and Adaptations of Machine Learning in Big Data—Applications in Real Cases in Agriculture
Round 1
Reviewer 1 Report
Kindly find attached
Comments for author File: 
Comments.pdf
Author Response
Dear Reviewer We appreciate your review and the time taken for it. We attach a file with the responses to each comment.
Author Response File: Author Response.pdf
Reviewer 2 Report
Dear author,
I have read this paper with interest and I hope you are open for some comments that may clarify some to even improve this presentation a little bit.
In the manuscript you showed very interesting research. The analysis of the use and adaptations of Machine Learning in Big Data Applications in agriculture is very important.
Understanding this phenomenon can help to reduce problems in coordinating activities on agricultural production.
But in my opinion a few things are missing. The problem with this manuscript is several inaccuracies, which are described in review.
You studied literature well. But I think you should supplement it. I especially think it should be complemented, especially with the theoretical content in Background (it is poor) in the context of agriculture .
There are also stylistic and technical errors. Some errors need to be corrected.
The scientific style avoids personal forms.
So, the manuscript can be published after summarizing the comments and suggestions contained in this review and approval by the editor.
Abstract
The authors should review the structure of the abstract, which is incomplete. There are no short conclusions.
Introduction
I don't really understand the first sentence - the population growth is related to food shortages?
The availability of food determines the size of the population (the world of animals and plants is characterized by this). This is a fact. But is it otherwise? The point is probably that the amount of food can determine the population. Please verify this.
L.24-26 Please specify the source because if this is [2] the data is very old. Forecasts quickly become obsolete.
L.28-29 Please state the source - who defines it?
L.63-64 Can the institute state? Maybe: In research by McKinsey Global Institute
[16] is wrongly quoted. Not "G.M. Date" but should be K. Slavakis, G. B. Giannakis and G. Mateos, "Modeling and Optimization for Big Data Analytics: (Statistical) learning tools for our era of data deluge," in IEEE Signal Processing Magazine, vol. 31, no. 5, pp. 18-31, Sept. 2014, doi: 10.1109/MSP.2014.2327238.
Background
Tables may have a footer – references?
L.110-111 - add "First"; 111-112 “Second”
In the paragraph above (110-114) there is a description of three dimensions. Here (115-116) you refer to four. Please link or explain it somehow
L.126-128 please indicate the reference (It is considered)
L.146-151 - is this paragraph necessary? Please do not write what will be in the chapter. This is an introduction to the chapter so you should change the style - describe why it is important; it should be put in problematic terms, explaining it.
Please change also in other chapter introductions (e.g. 5, 6, 7)
L157 - Why is weed detection 'another important problem' if you mention it in the paragraph above (154)? What about the remaining problems?
Describe Figure 2? What is the unit? The figure is unreadable. It is also worth to sort the data from the longest bar to the shortest one ;)
L172 “shows its evolution”
The scientific style avoids personal forms.
192-194. It seems to me that, referring to this research, the author [6] should be named after him. Besides, a few people and in line 198 you write "The author concludes ...". - what author? The authors.
192-204 This is the authors' opinion [6]. And what do you think? What is your opinion?
223-225. Please write more of these publications. Maybe a small discussion?
226. Rathor and Gyanchandani [37] and Divya, Bhargavi and Jyothi [38] presented a comparative analysis …
Likewise in lines: 528, 531, 537, 542, 544, 546, 553, 557, 559,
229. “So far there” Is this a good statement for publication form 2017 and 2018?
Table 4 unreadable
Research Methodology
257. What kind of research? The research presented in this manuscript.
276-277. Why in brackets? Please explain.
281-282. Why without Web of Science?
345-346. Why do 34 of 68 articles remain? This has not been explained.
Analysis
Please expand the introduction to the chapter.
I think the big tables should be in the attachments.
387-399. The dot should be at the end of the sentence.
Figure 3: Sort. Is it necessary to have the category "Farmers' insurance and finance" if it is 0?
Figure 10 - not described in the text. Please describe it.
524 - which?
569 Is this about figure 15?
Figure 15 - no description in the text. Wrong numbering of figures? Please check.
575-576. Please describe this figure 16. What does that mean?
Figure 16. Sort. Unreadable figure
Discussions
Change the introductory style.
Figure 17. Describe it. What does this mean?
Figure 18. Unreadable. Describe it. What does this mean?
And finally.
- Check and correct references (example [16]). Enter all required data.
- Check the numbering of references, tables and figures.
- Remove double spaces (eg. l.537).
Author Response
Dear Reviewer We appreciate your review and the time taken for it. We attach a file with the responses to each comment.
Author Response File: Author Response.pdf
Reviewer 3 Report
These are my main comments on the MS electronics-1075795) entitled:“ Use and adaptations of Machine Learning in Big Data Applications in real cases in Agriculture”
It is a thorough and extensive review focusing on digital and smart applications in the agricultural practice. I am also confident that Agriculture will be greatly changed in the very near future with the development of such technologies. I found the review very interesting but very hard to read, especially for readers “based” on the “agricultural side” and not on the “technological” like the authors.
My main objections concern the selection of sources for writing this review. Authors should try harder to locate original sources and avoid re-presenting older reviews. They have selected 34 papers (It was impossible for me to follow the selection procedure) but the “real cases” were vaguely presented. In my opinion the review should be re-written with the contribution of an agriculture expert who would be able to present practically the cases (e.g. irrigation, soil mapping, pest control, disease detection, fertilization, spraying etc).
Although authors have included a plethora of references in their study, most of them are reviews (almost half of them) and not original sources of experimental data. Most of all, there is extensive use of text (similar phrases, almost identical sentences) and figures from recent similar reviews (like 3, 4, 44 etc). Authors should emphasize on the elements that their study differentiates from recent similar reviews.
Many studies have been published during the last 5-6 years about the use of ML and algorithms on monitoring and detection of pest and diseases, on crop irrigation and other vital agricultural practices. Authors have ignored them. I strongly suggest to them to make a simple search on Scopus and locate some very nice and original studies with very interesting results that would greatly improve their review. There is no meaning presenting conclusions and figures from other reviews.
My proposal is to reject it for publication in "Electronics" and re-submit after making extensive changes according to the above.
Author Response
Dear Reviewer
We appreciate your review and the time taken for it. We attach a file with the responses to each comment.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
The authors have appropriately revised the manuscript
Author Response
We appreciate the reviewer's comment.
We have improved the English of the paper for a better understanding.
Reviewer 3 Report
After reading author responses to my comments my opinion remains the same. It seems that authors do not want or cannot re-write their review following my suggestions.
As a result, the title does not represent the text. It is impossible for an "agriculture scientist" to understand this paper.
However, the Editor may decide if this review is suitable for Electronics.
Author Response
We appreciate the reviewer's comment.
As suggested by the Editor of the journal, we have incorporated the following text in the abstract:
"This review may be of interest to computer or data scientists and electronic or software engineers."
We have also incorporated the following text in the Introduction:
"This review may be of interest to industry professionals, specifically to computer or data scientists and electronic or software engineers, who wish to get an updated view of the extent to which ML and Big Data have been applied and validated in agriculture."
On the other hand we have improved the English of the paper for a better understanding.
