The Estimation of the Long-Term Agricultural Output with a Robust Machine Learning Prediction Model

Round 1

Reviewer 1 Report

The manuscript draft is devoted to an interesting problem that touches on Long-Term Agricultural Forecasting approaches. The authors propose the marriage in honey bee optimization/ support vector regression model to minimize the effect of highly volatile data (outliers) and enhance the prediction accuracy. The results showed a better model adequacy than the comparison models.  Furthermore, the proposed model predicts long-term agricultural output more accurately and achieves higher directional symmetry than the other models.

Paper has practical value and has the scientific novelty of this paper. It has a logical structure. The paper is well-written and technically sound. The experimental section is good.

The proposed approach is logical, results are clear.

 

I have the following remarks:

1.      The situation in the world, in particular in agriculture, has changed significantly in recent times. The authors should give a detailed analysis of the situation and explore how modern circumstances affect agriculture and food security.

2.      The references may need to be updated.

3.      The authors mentioned that the datasets include the annual total agricultural output data from 1998 to 2008 and 2009 to 2022. But Fig. 1 presented data from 1998 to 2020. The authors should give all the data.

4.      The input data is very generalized. When forecasting such data, there are often problems with the adequacy of models and the accuracy of forecasts. The practical use of such data is also impractical. The authors should explain the purpose for which such data are used.

5.      Machine learning methods allow you to build a large number of models for forecasting, which have different characteristics of adequacy and efficiency of models. The authors should analyze the feasibility of building other models of machine learning.

6.      The authors should compare the results of the forecast with the latest real data. Such a comparison will demonstrate the value of the study and prove the possibility of using the built model for policy design.

7.      The conclusion section should be extended using: the limitations of the proposed approach; prospects for future research.

8.      Discussion may need to focus on what was achieved from this research and compare it with the literature.

 

 

Author Response

We thank the reviewer very much for their effort in reviewing our paper entitled “The Estimation of the Long-Term Agricultural Output with a Robust Machine Learning Prediction Model.” We also thank the reviewer for providing invaluable suggestions and comments, which proved very useful in revising the manuscript. The point-by-point responses to the reviewer’ comments are in the file.

Author Response File: Author Response.docx

Reviewer 2 Report

Dear authors,

The paper “The Estimation of the Long-Term Agricultural Output with a Robust Machine Learning Prediction Model” investigated the performance of the MBO/SVR model by using the annual total agricultural output in Taiwan. Overall, results of this study could provide valuable guidance for local forestry management. In my point of view, this manuscript need some minor clarifications.

Some general comments as fallow:

Line 42: it was mentioned that recently --> however, the references are not for recent studies except one. Update the references.

Table: Maybe it is better to show the changes as a graph to understand better.

Line 125-125: I suppose it is useless.

Any other studies use MBO algorithm? If Yes, please cite in introduction.

Step 2.2.2 and Performance measures: Some symbols are not readable.

Author Response

We thank the reviewer very much for their effort in reviewing our paper entitled “The Estimation of the Long-Term Agricultural Output with a Robust Machine Learning Prediction Model.” We also thank the reviewer for providing invaluable suggestions and comments, which proved very useful in revising the manuscript. The point-by-point responses to the reviewer’ comments are in the file.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Thank you for your interesting research.

The paper could be accepted in present form