Precise Recommendation Method of Suitable Planting Areas of Maize Varieties Based on Knowledge Graph
Jerzy Księżak
Round 1
Reviewer 1 Report
Paper title: Precise Recommendation Method of Suitable Planting Areas of 2 Maize Varieties
Based on Knowledge Graph
This research paper proposes a precise recommendation method for suitable planting areas of
maize varieties based on a knowledge graph and addresses the need for precision in promoting
new maize varieties and exploiting their potential in order to win the market competition. The
authors used the meteorology knowledge graph of maize ecological regions that is constructed
at the county level, and a RippleNet recommendation model to mine the potential spatial
correlation of maize variety suitability in different meteorological environments. This study
has a good piece of work that provides a data-driven solution for precise recommendations and
market positioning of maize varieties. The manuscript has a significant research contribution
and could be considered in the Agriculture journal after revision. Comments are as follows.
1. Explain the methodology used in the study for recommending suitable planting areas
for maize varieties?
2. How does the use of a knowledge graph improve the precision of the recommendation
method?
Introduction
? State the research gap (hypothesis) for undertaking the present investigation with
reference at the end of the introduction section.
Methods
? Improves the quality of Figure 1.
Results
? Written well
Discussion
? Discussion section lacks a comparison of the manuscript findings with
previous studies at different levels. Thus, improve it by adding more studies
section-wise.
Conclusion
? Written well
Author Response
Revision of my manuscript:
(1) According to the comments of the reviewers, I found that the chapter layout of my manuscript is not conducive to readers' reading, so I adjusted the paragraph layout of the manuscript. The contents of the adjustment are mainly concentrated in chapter 2. Materials and Methods.
1) The chapters in 2. Materials and Methods have changed from the original three sections to the current five summaries, as shown in the following table.
|
Original manuscript |
Revised manuscript |
|
2.1. The General Situation of Research Area |
2.1. The General Situation of Research Area |
|
2.2. Acquisition and Processing of Data |
2.2. Technical Process of this Research |
|
2.2.1. Test Data 2.2.2. Processing of test data |
2.3. Acquisition and Processing of Data |
|
2.3.1. Primitive Meteorological Data |
|
|
2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials |
|
|
2.3.3. Processing of Primitive Meteorological Data and Test Data |
|
|
2.3.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.4. Construction of Meteorology KG |
|
2.4.1. Schema Design of the Meteorology KG |
|
|
2.4.2. Storage of the Meteorology KG |
|
|
2.3.1. Construction of meteorology KG 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.5.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
|
2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet |
|
|
2.5.2.Concrete Implementation of Recommendation Model Based on RippleNet |
2) Added secondary title 2.2. Technical Process of this Research
The description of Figure 4 and Figure 4 of 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is proposed as 2.2. Technical Process of this Research;
3) The 2.2.1. Test Data in the original manuscript is adjusted to 2.3.1. Primitive Metrological Data and 2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials;
4) 2.2.2. Processing of test data in the original manuscript is adjusted to 2.3.3. Processing of Primitive Metrological Data and Test Data in the current manuscript;
5) The 2.3.1. Construction of Meteorology KG in the original manuscript is adjusted to 2.4. Construction of Meteorology KG, and the four-level titles under chapter 2.3.1. are adjusted to three-level title 2.4.1. Schema Design of the Meteorology KG and 2.4.2. Storage of the Metrology KG;
6) The 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is adjusted to 2.5. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet. The following contents of chapter 2.3.2. are adjusted to 2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet and 2.5.2. Concrete Implementation of Recommendation Model Based on RippleNet;
(2) Introduction: Add references to solve the problem of text source;
(3) Adjust the Figure 1 and Figure 2 in the original manuscript to solve the problem that the picture is not standard enough and the text in the picture is too small for reading;
(4) Change the three-level titles under Section 2.3. Acquisition and Processing of Data to make the titles directly reflect the content;
(5) Table 8 has been changed to solve the problem that is easy to cause misunderstanding by readers;
(6) Revised the titles of Table 9;
(7) Remove the five-level titles from the four-level title (The recommendation model based on RippleNet) of Section 2.5.2. Concrete Implementation Recommendation Model Based on RippleNet;
(8) Modify the content in the Table Algorithm 1 to accurately reflect the algorithm of the recommendation model;
(9) In view of the relatively high duplication rate of my manuscript, the contents of this manuscript were slightly modified to reduce the duplication rate of manuscripts.
Author Response File: 
Author Response.docx
Reviewer 2 Report
The manuscript will bring important scientific contributions.
Author Response
Dear reviewer：
The co-anthors and I would like to thank you for the time and effort spent in reviewing the manuscript. Thank you for your valuable suggestions, which are very helpful to improve my manuscript.
Revision of my manuscript:
(1) According to the comments of the reviewers, I found that the chapter layout of my manuscript is not conducive to readers' reading, so I adjusted the paragraph layout of the manuscript. The contents of the adjustment are mainly concentrated in chapter 2. Materials and Methods.
1) The chapters in 2. Materials and Methods have changed from the original three sections to the current five summaries, as shown in the following table.
|
Original manuscript |
Revised manuscript |
|
2.1. The General Situation of Research Area |
2.1. The General Situation of Research Area |
|
2.2. Acquisition and Processing of Data |
2.2. Technical Process of this Research |
|
2.2.1. Test Data 2.2.2. Processing of test data |
2.3. Acquisition and Processing of Data |
|
2.3.1. Primitive Meteorological Data |
|
|
2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials |
|
|
2.3.3. Processing of Primitive Meteorological Data and Test Data |
|
|
2.3.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.4. Construction of Meteorology KG |
|
2.4.1. Schema Design of the Meteorology KG |
|
|
2.4.2. Storage of the Meteorology KG |
|
|
2.3.1. Construction of meteorology KG 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.5.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
|
2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet |
|
|
2.5.2.Concrete Implementation of Recommendation Model Based on RippleNet |
2) Added secondary title 2.2. Technical Process of this Research
The description of Figure 4 and Figure 4 of 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is proposed as 2.2. Technical Process of this Research;
3) The 2.2.1. Test Data in the original manuscript is adjusted to 2.3.1. Primitive Metrological Data and 2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials;
4) 2.2.2. Processing of test data in the original manuscript is adjusted to 2.3.3. Processing of Primitive Metrological Data and Test Data in the current manuscript;
5) The 2.3.1. Construction of Meteorology KG in the original manuscript is adjusted to 2.4. Construction of Meteorology KG, and the four-level titles under chapter 2.3.1. are adjusted to three-level title 2.4.1. Schema Design of the Meteorology KG and 2.4.2. Storage of the Metrology KG;
6) The 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is adjusted to 2.5. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet. The following contents of chapter 2.3.2. are adjusted to 2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet and 2.5.2. Concrete Implementation of Recommendation Model Based on RippleNet;
(2) Introduction: Add references to solve the problem of text source;
(3) Adjust the Figure 1 and Figure 2 in the original manuscript to solve the problem that the picture is not standard enough and the text in the picture is too small for reading;
(4) Change the three-level titles under Section 2.3. Acquisition and Processing of Data to make the titles directly reflect the content;
(5) Table 8 has been changed to solve the problem that is easy to cause misunderstanding by readers;
(6) Revised the titles of Table 9;
(7) Remove the five-level titles from the four-level title (The recommendation model based on RippleNet) of Section 2.5.2. Concrete Implementation Recommendation Model Based on RippleNet;
(8) Modify the content in the Table Algorithm 1 to accurately reflect the algorithm of the recommendation model;
(9) In view of the relatively high duplication rate of my manuscript, the contents of this manuscript were slightly modified to reduce the duplication rate of manuscripts.
Reviewer 3 Report
The manuscript is well written. The meteorology, RippleNet model/algorithm is well explained in the spatial correlation of maize variety suitability in different meteorological environments.
As such the manuscript can be accepted.
Author Response
Dear reviewer：
The co-anthors and I would like to thank you for the time and effort spent in reviewing the manuscript. Thank you for your valuable suggestions, which are very helpful to improve my manuscript.
Revision of my manuscript:
(1) According to the comments of the reviewers, I found that the chapter layout of my manuscript is not conducive to readers' reading, so I adjusted the paragraph layout of the manuscript. The contents of the adjustment are mainly concentrated in chapter 2. Materials and Methods.
1) The chapters in 2. Materials and Methods have changed from the original three sections to the current five summaries, as shown in the following table.
|
Original manuscript |
Revised manuscript |
|
2.1. The General Situation of Research Area |
2.1. The General Situation of Research Area |
|
2.2. Acquisition and Processing of Data |
2.2. Technical Process of this Research |
|
2.2.1. Test Data 2.2.2. Processing of test data |
2.3. Acquisition and Processing of Data |
|
2.3.1. Primitive Meteorological Data |
|
|
2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials |
|
|
2.3.3. Processing of Primitive Meteorological Data and Test Data |
|
|
2.3.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.4. Construction of Meteorology KG |
|
2.4.1. Schema Design of the Meteorology KG |
|
|
2.4.2. Storage of the Meteorology KG |
|
|
2.3.1. Construction of meteorology KG 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.5.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
|
2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet |
|
|
2.5.2.Concrete Implementation of Recommendation Model Based on RippleNet |
2) Added secondary title 2.2. Technical Process of this Research
The description of Figure 4 and Figure 4 of 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is proposed as 2.2. Technical Process of this Research;
3) The 2.2.1. Test Data in the original manuscript is adjusted to 2.3.1. Primitive Metrological Data and 2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials;
4) 2.2.2. Processing of test data in the original manuscript is adjusted to 2.3.3. Processing of Primitive Metrological Data and Test Data in the current manuscript;
5) The 2.3.1. Construction of Meteorology KG in the original manuscript is adjusted to 2.4. Construction of Meteorology KG, and the four-level titles under chapter 2.3.1. are adjusted to three-level title 2.4.1. Schema Design of the Meteorology KG and 2.4.2. Storage of the Metrology KG;
6) The 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is adjusted to 2.5. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet. The following contents of chapter 2.3.2. are adjusted to 2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet and 2.5.2. Concrete Implementation of Recommendation Model Based on RippleNet;
(2) Introduction: Add references to solve the problem of text source;
(3) Adjust the Figure 1 and Figure 2 in the original manuscript to solve the problem that the picture is not standard enough and the text in the picture is too small for reading;
(4) Change the three-level titles under Section 2.3. Acquisition and Processing of Data to make the titles directly reflect the content;
(5) Table 8 has been changed to solve the problem that is easy to cause misunderstanding by readers;
(6) Revised the titles of Table 9;
(7) Remove the five-level titles from the four-level title (The recommendation model based on RippleNet) of Section 2.5.2. Concrete Implementation Recommendation Model Based on RippleNet;
(8) Modify the content in the Table Algorithm 1 to accurately reflect the algorithm of the recommendation model;
(9) In view of the relatively high duplication rate of my manuscript, the contents of this manuscript were slightly modified to reduce the duplication rate of manuscripts.
Reviewer 4 Report
The studied corn parameters and meteorological indicators were neglected in the results and disscusion
Comments for author File: 
Comments.pdf
Author Response
Dear reviewer：
The co-anthors and I would like to thank you for the time and effort spent in reviewing the manuscript. Thank you for your valuable suggestions, which are very helpful to improve my manuscript.
Revision of my manuscript:
(1) According to the comments of the reviewers, I found that the chapter layout of my manuscript is not conducive to readers' reading, so I adjusted the paragraph layout of the manuscript. The contents of the adjustment are mainly concentrated in chapter 2. Materials and Methods.
1) The chapters in 2. Materials and Methods have changed from the original three sections to the current five summaries, as shown in the following table.
|
Original manuscript |
Revised manuscript |
|
2.1. The General Situation of Research Area |
2.1. The General Situation of Research Area |
|
2.2. Acquisition and Processing of Data |
2.2. Technical Process of this Research |
|
2.2.1. Test Data 2.2.2. Processing of test data |
2.3. Acquisition and Processing of Data |
|
2.3.1. Primitive Meteorological Data |
|
|
2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials |
|
|
2.3.3. Processing of Primitive Meteorological Data and Test Data |
|
|
2.3.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.4. Construction of Meteorology KG |
|
2.4.1. Schema Design of the Meteorology KG |
|
|
2.4.2. Storage of the Meteorology KG |
|
|
2.3.1. Construction of meteorology KG 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.5.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
|
2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet |
|
|
2.5.2.Concrete Implementation of Recommendation Model Based on RippleNet |
2) Added secondary title 2.2. Technical Process of this Research
The description of Figure 4 and Figure 4 of 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is proposed as 2.2. Technical Process of this Research;
3) The 2.2.1. Test Data in the original manuscript is adjusted to 2.3.1. Primitive Metrological Data and 2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials;
4) 2.2.2. Processing of test data in the original manuscript is adjusted to 2.3.3. Processing of Primitive Metrological Data and Test Data in the current manuscript;
5) The 2.3.1. Construction of Meteorology KG in the original manuscript is adjusted to 2.4. Construction of Meteorology KG, and the four-level titles under chapter 2.3.1. are adjusted to three-level title 2.4.1. Schema Design of the Meteorology KG and 2.4.2. Storage of the Metrology KG;
6) The 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is adjusted to 2.5. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet. The following contents of chapter 2.3.2. are adjusted to 2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet and 2.5.2. Concrete Implementation of Recommendation Model Based on RippleNet;
(2) Introduction: Add references to solve the problem of text source;
(3) Adjust the Figure 1 and Figure 2 in the original manuscript to solve the problem that the picture is not standard enough and the text in the picture is too small for reading;
(4) Change the three-level titles under Section 2.3. Acquisition and Processing of Data to make the titles directly reflect the content;
(5) Table 8 has been changed to solve the problem that is easy to cause misunderstanding by readers;
(6) Revised the titles of Table 9;
(7) Remove the five-level titles from the four-level title (The recommendation model based on RippleNet) of Section 2.5.2. Concrete Implementation Recommendation Model Based on RippleNet;
(8) Modify the content in the Table Algorithm 1 to accurately reflect the algorithm of the recommendation model;
(9) In view of the relatively high duplication rate of my manuscript, the contents of this manuscript were slightly modified to reduce the duplication rate of manuscripts.
Author Response File: Author Response.docx
Reviewer 5 Report
- the manuscript is an important supplement to the research regarding maize,
- the manuscript is a valuable supplement to research carried out so far in other research centers,
- the obtained results will be an indication for farmers engaged in the breeding of ruminant and monogastric animals, -
the results presented will be a guide to corn growers
- all papers are cited in the manuscript
- important is the conclusion that it is:
- this model was compared with six other traditional machine learning methods to assess the performance of the model.
- in addition, the proposed model was compared with the graph attention neural network (GAT) [34] model and the graph convolution neural network (GCN) [35] model.
- the performance of the proposed model was additionally assessed using the following five evaluation indicators: Accuracy rate, Precision rate, Recall rate, F1-Score and Area under Curve (AUC).
- in addition, compared with GAT (69.4%) and GCN (72.4%), RippleNet also performed better in terms of accuracy, but its recall value was slightly lower than the GCN model and better than the GAT model. RippleNet also produced good precision (78.8%), F1 (83.9%) and AUC (80.7%) scores, which were considerably higher than in other methods, further verifying the superiority of the model. The constructed meteorology KG of the Huang-Huai-Hai ecological region presented in this paper will help to precisely locate the suitable planting areas of maize varieties from the ecological region-level to the county-level and connect all counties in the Huang Huai-Hai ecological region.
- the RippleNet recommendation model was used to construct a propagation network of preference for variety environmental adaptability, which also explored the implicit relationship between the variety field performance and the meteorological factors in the planting area. Consequently, a precise evaluation of the suitability of maize varieties was achieved.
- I don't feel qualified to judge about the English language and style
Author Response
Dear reviewer：
The co-anthors and I would like to thank you for the time and effort spent in reviewing the manuscript. Thank you for your valuable suggestions, which are very helpful to improve my manuscript.
Revision of my manuscript:
(1) According to the comments of the reviewers, I found that the chapter layout of my manuscript is not conducive to readers' reading, so I adjusted the paragraph layout of the manuscript. The contents of the adjustment are mainly concentrated in chapter 2. Materials and Methods.
1) The chapters in 2. Materials and Methods have changed from the original three sections to the current five summaries, as shown in the following table.
|
Original manuscript |
Revised manuscript |
|
2.1. The General Situation of Research Area |
2.1. The General Situation of Research Area |
|
2.2. Acquisition and Processing of Data |
2.2. Technical Process of this Research |
|
2.2.1. Test Data 2.2.2. Processing of test data |
2.3. Acquisition and Processing of Data |
|
2.3.1. Primitive Meteorological Data |
|
|
2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials |
|
|
2.3.3. Processing of Primitive Meteorological Data and Test Data |
|
|
2.3.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.4. Construction of Meteorology KG |
|
2.4.1. Schema Design of the Meteorology KG |
|
|
2.4.2. Storage of the Meteorology KG |
|
|
2.3.1. Construction of meteorology KG 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
2.5.Recommendation Model of Suitable Planting Area for Maize Based on RippleNet |
|
2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet |
|
|
2.5.2.Concrete Implementation of Recommendation Model Based on RippleNet |
2) Added secondary title 2.2. Technical Process of this Research
The description of Figure 4 and Figure 4 of 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is proposed as 2.2. Technical Process of this Research;
3) The 2.2.1. Test Data in the original manuscript is adjusted to 2.3.1. Primitive Metrological Data and 2.3.2. Primitive Test Data from Chinese National Maize Variety Field Trials;
4) 2.2.2. Processing of test data in the original manuscript is adjusted to 2.3.3. Processing of Primitive Metrological Data and Test Data in the current manuscript;
5) The 2.3.1. Construction of Meteorology KG in the original manuscript is adjusted to 2.4. Construction of Meteorology KG, and the four-level titles under chapter 2.3.1. are adjusted to three-level title 2.4.1. Schema Design of the Meteorology KG and 2.4.2. Storage of the Metrology KG;
6) The 2.3.2. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet in the original manuscript is adjusted to 2.5. Recommendation Model of Suitable Planting Area for Maize Based on RippleNet. The following contents of chapter 2.3.2. are adjusted to 2.5.1. The Role of Knowledge Graph in Recommendation Model Based on RippleNet and 2.5.2. Concrete Implementation of Recommendation Model Based on RippleNet;
(2) Introduction: Add references to solve the problem of text source;
(3) Adjust the Figure 1 and Figure 2 in the original manuscript to solve the problem that the picture is not standard enough and the text in the picture is too small for reading;
(4) Change the three-level titles under Section 2.3. Acquisition and Processing of Data to make the titles directly reflect the content;
(5) Table 8 has been changed to solve the problem that is easy to cause misunderstanding by readers;
(6) Revised the titles of Table 9;
(7) Remove the five-level titles from the four-level title (The recommendation model based on RippleNet) of Section 2.5.2. Concrete Implementation Recommendation Model Based on RippleNet;
(8) Modify the content in the Table Algorithm 1 to accurately reflect the algorithm of the recommendation model;
(9) In view of the relatively high duplication rate of my manuscript, the contents of this manuscript were slightly modified to reduce the duplication rate of manuscripts.
Round 2
Reviewer 4 Report
Now it is OK
