Online User Review Analysis for Product Evaluation and Improvement

Round 1

Reviewer 1 Report

The paper improved from its earlier version. Thank you for the corrections. Some minor corrections:
Please check all the abbreviation. Some are missing like B2C, C2C.
It will be better to include the full notation of S (Satisfaction), A (Attention), and P (Priority) in Table 7 and Table 8 to understand it better.
Average positive emotion value and Satisfaction are both indicated by S whereas product attribute and Attention are indicated by A. These are confusing to understand the method. Please revise these terms. 

Author Response

Thanks for your comments and all modifications are highlighted.

 

Point 1: Please check all the abbreviation. Some are missing like B2C, C2C.

Response 1: We have carefully checked all the abbreviations and added missed ones.

Line 95 RQ – Research Question

Line 205 NLP – Natural Language Processing

Line 219 B2C – Business to Customer

Line 220 C2C – Consumer to Consumer

Point 2: It will be better to include the full notation of S (Satisfaction), A (Attention), and P (Priority) in Table 7 and Table 8 to understand it better.

Response 2: Thanks for your suggestion and we have included the full notation in the two tables.

 

Point 3: Average positive emotion value and Satisfaction are both indicated by S whereas product attribute and Attention are indicated by A. These are confusing to understand the method. Please revise these terms.

Response 3: Sorry for the confusion. To eliminate the confusion, we have added more explanations for the three indicators (Line 176).

“We construct three indicators for product evaluation - User satisfaction (S), User Attention (A) and Priority (P). As the three indicators demonstrate concrete product attributes, each product attribute will have the three indicators. And to distinguish the overall indicators and concrete indicators for each product attributes, we add subscript for the abbreviations (e.g., for the 1-st product attribute, User satisfaction is marked as S₁”.

In addition, S is for user Satisfaction only and A is for user Attention only. It is our fault to annotate the abbreviations in the improper position and revisions are in Line 182 and Line 188.

Reviewer 2 Report

The authors have made a number of considerable modifications and I believe the manuscript is ready. 

Author Response

Thanks for your comments.

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

Dear authors

You have done an interesting job, however, I think you should adjust some sections to improve the existing work. 

On the one hand, I would advise you to add in the Keywords section as many keywords as the journal allows, this will improve your positioning in the various databases such as Web of Science or Scopus. On the other hand, you should include "Sentiment Analysis". 

Within the Introduction I would recommend you to add a subsection with Background or Theoretical Framework, do a little literature review of existing and recent works in online review...

Under the introduction you should add a subsection / or a full section with your Hypothesis or the Research Questions you hope to answer. For example:

H1/RQ1.....
H2/RQ2....
H3/RQ3....

The second section is confusing. I, would write Section 2. Materials and Methods and from there create subsections (Example 2.1, 2.2.) and sub-subsections. (Example 2.1.1, 2.2.1). 

At the end of the paper, I would try to enrich the conclusion better (and merge it Discussion and Conclusion). I miss more discussion and compared with recent works of other authors. You can check the Decision Support Systems and Electronic Commerce or Expert Systems to take some insight. 

In addition, you should add a subsection with "Future Research Lines". Let's keep the research line door open to other researchers to pursue new investigations.

I remain at your disposal. 
Best regards, 

 

Reviewer 2 Report

In this study, the authors build a product evaluation indicator system, and put forward improvement strategies for the product with opinion mining and sentiment analysis with online reviews. The topic is interesting, but I believe there is very little academic and theoretical contribution to this work. This is just a simple case study that may be suitable for a conference proceeding not for a journal publication.

 

1. It is really unclear where the scholarship and academic challenge lies i.e. the rationale for conducting this research is not clear. There is a lack of scholarship, justification for the research, or potential impact (at least detailing what it might be, would have been good).
2. Discuss the research questions more elaborately in the introduction.
3. What is the objective of this study?
4. The research gap of the earlier studies needs to be explained in more detail. The work doesn't present a suitable enough niche to show a contribution to a gap that needs filling or indeed justify that that there is a gap or contribution to be made. Add a separate research gap and highlights sections in the introduction section.
5. The overall structure of this study needs major modification. It is very odd to mention the overall framework of this work (Figure 1) in introduction section. The authors should follow the papers from this journal to understand the format or structure.
6. Provide appropriate references for the attributes mentioned in Table 1.
7. “By consulting the opinions of professional smartphone designers….” What was the method of conducting these consultations? Who are those designers? How had you selected them? Are they expert enough to provide the judgement?
8. The authors used simple weighted method to perform the review usefulness analysis. I did not find academic contribution in this analysis.
9. Useful reviews are divided into 15 groups. Why 15 groups? Why not more or less?
10. The authors mentioned that R² indicates that the model has a good fitting effect with the data. Is only 0.537 R² enough for this study? Where is the reference of this benchmark?
11. The authors mentioned that an average error of 9.9% is small. How much error is large for this type of study? Where is the reference of this benchmark?
12. The conclusion section should be more precise in the case of the presentation of scientific contributions and limitations to the work done.

 

Reviewer 3 Report

I read carefully the paper entitled: ”Online User Review Analysis for Product Evaluation and Improvement”.

Although the paper has a certain scientific level, we notice some weaknesses: - On the whole, the work has a relatively simplistic character, the results being intuitive almost from the beginning. - Part required - Literature Review (1.5 pages), after Introduction. 

If they read carefully the text of the paper, the authors will detect for themselves the negligences ...

But he must apply the rules of the Journal.

It would be good to expand the bibliography with some articles from prestigious scientific journals (WoS) published in 2020 and 2021.