Identifying the Key Hazards behind Website Drop-Offs by Solving a Survival Problem

Round 1

Reviewer 1 Report

1- According to what measure was the efficiency of the proposed system determined? Please specify in the abstract.

2- As long as the two consecutive paragraphs are for the same reference, there is no need to mention it twice. As in the following paragraphs: "Rincon et. al (2022) have conducted a study to measure traffic and drop-offs on Colombian banking establishments [3]. Rincon et al. al have found that the websites of Colomstatement banking establishments were well positioned and presented low bounce rates and drop-offs [3].".

3- It is preferable to add a paragraph indicating the structure of the research paper at the end of the introduction section

4- No previous related work was mentioned to show the researcher's contributions in this field and the problems he overcame

5-The references from which the equations were extracted were not indicated

6-The mentioned references are very few and it is possible to add at least five recent references published during the last five years

7-The work was not compared to other works, and the tables and figures extracted for the results were not well discussed

 

Author Response

Thank you for the valuable feedback. Kindly refer to the attached which details the amendments made by the authors in response to your feedback.

Author Response File: Author Response.docx

Reviewer 2 Report

applsci-2443345

Identifying the key hazards behind website drop-offs using the traditional cox proportional hazard model and the modern machine learning survival random forest model

 

Generally, the manuscript is not well written to deny it for publication. However, I would like to propose several questions for the author to explain, so as to effectively improve the quality of this manuscript. Specific comments are as follows:

--Title should be more focused on which method you used.

--In abstract, verbose and tasteless talk about the background of the research is not good to make readers’ boiling. Please add more descriptions about the method you used in the Abstract.

-- Please given detail explanations about the traditional cox proportional hazard model as well as the modern machine learning.

--Please carefully check typos, for example, variables in formulas should be carefully checked.

--A flowchart to brief describe the method should be given.

--References should be added.

Author Response

Thank you for the valuable feedback. Kindly refer to the attached which details the amendments made by the authors in response to your feedback.

Author Response File: Author Response.docx

Reviewer 3 Report

Contribution/Summary: This study identified online drop-off hazards using survival analysis models, specifically the cox proportional hazard model and the survival random forest model. Google Analytics acquired data from an engineering services company's website. The study investigated why website visitors leave early. Survival random forest outperformed the cox proportional hazard model on non-linear features. This study may assist website owners and businesses enhance website design and user experience to reduce drop-offs and increase engagement.

Comments/Suggestions:
1. The paper's title emphasizes the study's major focus and significance, but its length may challenge readers. Long titles may make it tougher for readers to easily identify the study's core topic and find the paper in online searches. Thus, a more succinct title that truly reflects the paper's substance may be advantageous. Remove unnecessary words or use shorter substitutes for title terms.
  2. More information on the methods used to identify website drop-off hazards will improve the abstract. It would be great to explain how the cox proportional hazard model and survival random forest model were applied to Google Analytics data. This would assist readers comprehend the study's approach and assess its validity.   3. Short papers may not provide enough background or depth to clarify the research issue, methods, and findings. Readers may have trouble assessing the study's impact.    4. A limited number of references may indicate that the paper does not engage with the relevant literature, which may restrict its effect and relevance.   5. The research issue, which is the identification of major risks causing website drop-offs, is well-explained in the introduction. However, further information regarding the precise research question, the goals of the study, the methods used to analyze the data, and other pertinent topics would be beneficial.
6. The introduction might also be strengthened by including greater context regarding the significance of the study's findings and how they add to the body of knowledge on website drop-offs.

7. Given that Section 4 is much longer than the other sections, it could be beneficial to analyze its content and see whether it can be shortened or divided into more manageable sub-parts. This could make the paper flow more naturally and make it easier to follow by balancing the length of the various sections.   8. I recommend adding a paragraph on formal approaches for AI-based technique verification to this study to improve its quality and impact. Formal approaches, which utilize mathematical models and logic to check system correctness, are increasingly significant in AI-based technique development and validation. Please extend the paragraph you added and add more references.
9. Some relevant references related to this topic that the authors may want to consider include:

a.https://ieeexplore.ieee.org/abstract/document/9842406

b. https://incose.onlinelibrary.wiley.com/doi/abs/10.1002/inst.12434
  10. The paper might be made better by including a more thorough description of how the results will affect the website under study as well as the larger field of website design and marketing.
11. The writers might go into more detail about how to solve the hazards found in order to decrease drop-offs and boost page views. This could include suggestions for bettering website design, content optimization, or more precise marketing campaign targeting.

12. The study's limitations and possible research directions, such examining the effects of various marketing tactics or assessing the efficiency of particular website design elements, might also be discussed by the authors.

 13. Finally, the authors could discuss the significance of the study's findings in light of the existing literature and emphasize how the findings advance our knowledge of visitor behavior and website drop-offs. Overall, a more thorough and contextualized discussion of the results' implications could serve to strengthen the paper's conclusion and increase its overall effect.

Can be improved.

 

Author Response

Thank you for the valuable feedback. Kindly refer to the attached which details the amendments made by the authors in response to your feedback.

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Well revison to support it for publication as it is.

Reviewer 3 Report

The authors considered all my comments and suggestions. Good luck.