AWdpCNER: Automated Wdp Chinese Named Entity Recognition from Wheat Diseases and Pests Text

Round 1

Reviewer 1 Report

1. It is important to understand how the authors obtained these 0.88 percentage points, 0.44 percentage points, and 0.75 percentage points. There is a need for some validation of these percentages.

2. The lines from 73 to 74 of the paper which are the main part of the paper, are not clear. Ensure that your statement is clear and related to the literature you are citing.

3. This should be the blockdiagram of the proposed method. It needs to be made more clear in figure 1.

4. I think Figure 3 is fine, but it should also provide a image identification as well.

5. I would like to know where the equation is implemented in your proposed method.

6. There is a need for a more specific name for Figure 4 in addition to the model name.

7. It is not clear to me what table 3 means.

8. There are a number of equations in this document that are not in the standard equation format.

9. What is the formula for getting the percentage F1 score in Table 4, Table 5 and Table 6? 

10. There needs to be some results in the form of images in the paper

11. Figure 5 has a very long name. Please shorten it and make it more specific.

 Minor editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

AWdpCNER：Automated Wdp Chinese Named Entity Recognition from Wheat Diseases and Pests Text

1. Very interesting research entitled “AWdpCNER：Automated Wdp Chinese Named Entity Recognition from Wheat Diseases and Pests Text”.

2. Correct the structure of the article (see attached file).

3. I suggest eliminating the paragraph on lines 97-102. It is not necessary to comment on what will be covered in later sections.

4. Delete the texts that are in Chinese. Lines: 152, 155, 156, 159, 160, 278, (table 3), 356 and 357.  

5. On line 250 it says: “B-lable1 followed by I-lable2.“.  Maybe I should say: “B-label1 followed by I-label2.”. Check please.

6. Delete the information in Chinese from table 3. All information must be presented in English.

7. I suggest developing an algorithm of the "AWdpCNER model". I suggest that the algorithms in this article use the following format: (see attached file).

8. The title of figure 5 should be short. The figure should be explained in the previous paragraph.

9. The results obtained are not clear, please explain them adequately.

10. Consider future work on this research.

11. Very good bibliography.

The article has good content and very interesting.

Authors are requested to make all indicated corrections.

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

This paper is concerned with named entity recognition applied in the field of agricultural pests and diseases for wheat, in the Chinese texts. The authors have also developed a corpus from authoritative text sources for this purpose, collected automatically with crawlers and OCRs, then cleaned manually. Data preprocessing and augmentation is also involved, after selecting 21 types of entities.  The training was based on the ALBERT model, a BiLSTM layer, a CRF layer and further rules amendment. The experiments show overall performance metrics to be very good: precision is 94.76%, the recall is 95.64%, and the F1-score is 95.29%.

Some comments on the paper: Can you emphasise your results in comparison with state of the art (I find only one paper with a metric, it would be good to see more). I would be useful to specify also the description of the dataset after augmentation. Perhaps in the discussion section you can mention what could be the limitations of the model, if any, and more details on the significance of the results, which are very good. The conclusions are numbered, I believe that is a mistake. Please also check phrasing, try no to repeat words in the same sentence if possible (ex. page 2 line 47). Great work and a lot of effort put into it!

Please also check phrasing, try no to repeat words in the same sentence if possible (ex. page 2 line 47). 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Despite the fact that the authors responded to all our comments, I think that some image results might be needed to better understand the comparison of Figure 5.

Moderate editing of English language required

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

I thank the authors for having made all the observations.

Author Response

Please see the attachment.

Author Response File: Author Response.docx