A Robust Chinese Named Entity Recognition Method Based on Integrating Dual-Layer Features and CSBERT

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Thank you for addressing such an interesting problem in Chinese Named Entity Recognition, but we have some remarks and questions that may help to highlight the work done:

1. There is an honorific system in Asian languages. It also includes suffixes and prefixes. Do these factors affect the accuracy of the model when using morphemes in natural language analysis?

2. Can you provide more information about the dataset used to train the model? How much data did you learn in total? Please describe your data succinctly.

3. In Table 4.1, the quantity of entities by entity category used for model learning is unbalanced. Why did we train unbalanced? Why didn't we train in balance?

4. Please explain why precision, recall, and F1 differ depending on the entity type in Table 4.4.

 

 

 

 

Author Response

We sincerely thank the editor and all reviewers for your valuable feedback that we have used to improve the quality of our manuscript.

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

This paper proposes CSBERT-IDCNN-BiLSTM-CRT to address the issues of lack of datasets and low accuracy in Chinese NER.

It is organized well, but it would be better if the authors consider and address the followings:

 

- NER needs to be stated with its full name at its first use in Introduction.

 

- More recent works (published in recent 3~4 years) need to be compared with the proposed model and dataset.

 

- Cite the references in which the models compared in Table 4.3 are proposed.

 

- There is only one dataset used for the evaluation. And there is an imbalance in the number of entities in the dataset. Is it reasonable to show P, R, and F1 by each entity in Table 4.4 in this situation?

 

- Further experiments on additional datasets accompanying comparison and analysis are needed to improve the quality of this paper.

Author Response

We sincerely thank the editor and all reviewers for your valuable feedback that we have used to improve the quality of our manuscript.

Please see the attachment.

Author Response File: Author Response.pdf