Towards Reliable Baselines for Document-Level Sentiment Analysis in the Czech and Slovak Languages

Round 1

Reviewer 1 Report

The idea of the paper is interesting, however, it needs substantial improvement before publication. Below are the comments that need to be addressed

1.     Abstract need to be modified, the author contribution is not clear

2.     The technical content is good however the motivation of the paper is not clear. Why is your proposal needed? What are the challenges involved? What solutions already exist for the problem you want to solve? What are their limitations and drawbacks?

3.     Add some latest references

4.     Add a comparison table at the end of literature review section to clearly identify the gaps and to show current state-of-the-art.

5.     Add few introductory lines between section 5 and 5.1

6.     The paper needs an English review. There are several misspelled words and undefined acronyms.

7.     The conclusion is very long, summarize the conclusion and add the remaining details in discussions section

8.     The heading of the conclusion section be “Conclusion and future work” as last paragraph I related to future work

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

First of all congratulation on submitting the manuscript to the MDPI journal. To improve the quality of the manuscript the following suggestions could be considered:
1) In the abstract of the manuscript I would suggest more focus on methods that have been used and on the novelty of this manuscript. Now it does not clear at all what has been done and obtained.
2) It is obvious that it is hard sometimes to understand text sentiment and it depends on the context, but anyway, how this sentence, for example, "This vacuum cleaner really sucks," (positive) can be positive at all? No one writes like this text to describe a good vacuum cleaner, or maybe I am wrong. Maybe at least the example could be less ambiguous.
3) Line 56, [7,6] it would be nice to follow the order of the numbers.
4) The structure of the manuscript should be presented at the end of the Introduction, not at the end of Related works.
5) At the end of the Related works authors could highlight what literature analysis shows up the problem and how the authors will solve that problem in the rest of the manuscript.
6) It is hard to understand the research when the dataset has not been described, at least a little bit should be presented.
7) I read the manuscript a few times but sorry, I do not understand anyway the purpose and the novelty of it, the text is hard to read, and the lack of focus on the main aspects. Everything is focused on other research and what has been done in this, is not highlighted. Maybe it is just my problem to understand it.
8) The a lot of minor mistakes in the manuscript, the authors should read the one more time the manuscript closely and fix the mistakes.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf