Attention–Survival Score: A Metric to Choose Better Keywords and Improve Visibility of Information

Round 1

Reviewer 1 Report

The article presents an algorithm that allows to "calculate" an alternative keyword list to the one provided by the authors of a research paper, so that the new list potentially increases the visibility of the paper.

The idea behind the paper is interesting and sufficiently innovative. However, the quality of the manuscript is too low for publication. The following two aspects should be significantly improved:

- The validation of the proposal, presented in the section entitled "Refinement algorithm validation" is vague. While the requirements for participation in the validation are listed, the characteristics of the final participants are not made explicit. Thus, it is not clear that they are optimal individuals for this role. Validators probably should be senior researchers. It also seems insufficient that the validators only read the title of the paper from which the starting keyword lists have been extracted. As described in the body of the article, the alternative keywords may be quite different from the original ones, which may not represent the content of the paper. It would also be interesting to provide several alternative keyword lists (calculated with different approaches, e.g. with different ontologies or different alpha) and evaluate their similarity to the original one.

- The manuscript should be reviewed in detail by a native English speaker. It is important to improve the narrative, so that it conforms to the formal model of research articles. Among other improvements: i) it is advisable to use the third person (except when commenting on personal appreciations); 2) it is not clear why hypotheses h1 and h2 are posed, since, although they are related to the stated objective, they do not seem to serve to demonstrate that the objective has really been achieved (the objective, I understand, is to achieve a set of keywords that offers more visibility to a paper); 3) section entitled "alpha parameter" should be substantially improved (and Figure 8 reworked), as it is not apparent how the alpha factor influences the results (Figure 8 should, perhaps, present the value of AS versus "alpha").

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Strength:

Overall, the proposed method in the paper seems interesting and useful for authors who want to improve the visibility of their papers. The idea of selecting alternative keywords with high Attention-Survival score, which combines the popularity and survival score of keywords, is unique and could potentially help authors find effective and less competitive keywords. Additionally, the use of ontologies to ensure the semantic relationship between alternative keywords and input keywords is a good approach to improve the quality of the suggested keywords. The paper also presents a thorough evaluation of the proposed method, which includes testing the sensibility of the ontology and conducting a survey for human validation. The positive results obtained from the tests indicate that the proposed method is effective and reliable.

Suggestions:

1.      However, without more information on the specific details of the algorithm and the datasets used, it is difficult to fully assess the paper's contributions and limitations.

2.      Additionally, it would be helpful to know if the proposed method has been implemented and tested in real-world scenarios.

3.      Rephrase your paper structure section to reflect better. You could consider:

Section 1 provides a literature review, which describes the current state-of-the-art and theory that underpins the paper. Section 2 details the Attention-Survival model, which presents our theoretical proposal. Section 3 provides information on our experimental design, including details about the dataset we used, an analysis of the WordNet and CSO ontologies, and examples that illustrate our refinement algorithm. Section 4 summarizes our conclusions, which are based on the results obtained from our experimental design. Section 5 contains the bibliography, which lists all the sources cited throughout the manuscript. Finally, Section 6 contains the appendix, which includes supplementary information that supports the main text of the manuscript.

4.      Some grammatical errors are still present in the manuscript. I recommend thorough proofreading of the manuscript.

Conclusion

Overall, the paper seems to be a valuable contribution to the field of information retrieval and could potentially benefit authors who want to improve the visibility of their papers.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors have improved the manuscript sufficiently.

Author Response

Dear reviewer,

Thank you for your valuable comments, which have been impacted positively the quality of our work.