Recommendation Algorithm Using SVD and Weight Point Rank (SVD-WPR)

Round 1

Reviewer 1 Report

 

The authors have proposed a recommendation algorithm that uses ranking-based collaborative filtering and matrix factorization method. The paper is well-structured. The authors started from reviewing the related research and on that basis they identified the research problem. The proposed approach using Singular Decomposition Value algorithm and Weight Point Rank algorithm is presented in Section 3. In Section 4 the authors presented the experimental results, which show the comparison of effectiveness and efficiency of the proposed algorithm and several other state-of-the-art approaches. The advantage of the proposed algorithm is particularly evident in the case of efficiency.

Please address the following issues:

1. What are the limits of applicability and limitations of the proposed algorithm?

2. Please clearly indicate in the Introduction the main contributions of the paper.

3. Which elements of the proposed approach are new/innovative, and which are taken from or based on other methods?

4. Is the proposed approach independent of the field of application? What should be eventually done to use it in other application domains?

5. Please improve the English language when it comes to grammar and style.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

One of the problems that mentions in this study is on how to get accurate recommendation ranking. However, due to the increasing number of items that is processed will make the execution time take more times. 

1.  The SVD and the WPR is provided in the manuscript. However, the reason why the authors choose this two methodologies should discuss in the introduction sections.

2. The research motivation is not discuss in details by the authors. The related works discussion should be able to determine the research position, including what the differences are between each previous state-of-the-art technique including the research motivation. Add some details with drawbacks from previous works to strengthen this work's contribution.

3. The experiments discuss on how well the algorithms is performed. Since this is a combined model. Add some experiments to show how the combined models is more effective compare with each methodology is used in the manuscript.

4. There are various recommendation algorithms in recent years. Add some state-of-the-arts models as comparison to define how well the algorithms that is proposed in this study in some aspects, i.e., time and number of data.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors already revised all the reviewers comments. However, there are some errors and please ensure your articles has been proofread before submitting this study.

Author Response

Thank you for your review and suggestion. We have re-read our article, corrected some errors, and proofread before submitting this article.