Comparing Hierarchical Approaches to Enhance Supervised Emotive Text Classification

Round 1

Reviewer 1 Report

In this paper, the authors mainly "provided a comparative investigation of how hierarchical classification methods which extend traditional evaluation metrics to consider the hierarchical classification scheme, can improve the performance of the classification of emotive texts, as well as classifying fine-grained emotions that are not limited to a finite number of classes." and the results of their experiments are very good. But there are also some shortcomings:

1. Evaluation metrics can be used to assess model performance, but how do their extended evaluation metrics improve the effect of the model? If the author provides the code of their experiments, it may help to eliminate the confusion of readers.

2. The article does not define the problem which can provides a clear focus for the research and help readers understand the objective of the study.

3. There are some errors in the article, for example, in "3.3. Mapping texts to impressions ", the author said "Figure 2 reports the top 20 most..." Actually, it is Figure 3. And the number of emotions shown in Figure 3 is 22, not 20.

4. In 5.1., the authors said "The Simple Logistic and Random Forest models failed to complete classifications after running for 2 days." but the authors did not provide an explanation for this occurrence.

     There are many long and complicated sentences in this paper, which makes it difficult for readers to understand the content of the paper. In addition, There are some unclear expressions in the paper. For example, in the abstract, the author said that "This paper provides a comparative investigation of how four extended evaluation metrics which consider the characteristics of the hierarchical scheme can be applied and subsequently improve the performance of the classification of emotive texts", but did not explain what "four extended evaluation metrics" mean in the abstract of the paper.

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

A good paper. See the attached pdf file for comments.

Comments for author File: Comments.pdf

Overall great. See the minor comments in the attached pdf file.

Author Response

Please see attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have revised the paper according to my comments. Now, the paper has improved a lot and I think it can be accepted.

After modifications, the readability of the paper has been improved.