Bayesian Model Selection for Addressing Cold-Start Problems in Partitioned Time Series Prediction

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This paper presents an innovative and approach for addressing the cold-start problem in time series analysis using Bayesian model selection. This study is interesting. However, there are still a few issues that should be discussed, as outlined below.

Comment 1) Model Construction: This approach allows for the integration of prior knowledge and statistical evidence, even when data are sparse or incomplete. It enables the selection of the most likely models that could have generated the observed data, improving the accuracy of time series predictions in scenarios with limited initial data. However, this method mainly reflects the evaluation of which Bayesian model works better under different parameter combinations as the best model. Is the selection criterion or principle mainly based on different distance measurement results at different significance levels? Can the author express it more prominently and clearly?

Comment 2) Model Comparison: In the simulation experiment results and application experiment results, I did not see any comparison and analysis results with machine learning or advanced collaborative filtering methods. Would the author consider adding them? Similarly, the advantages of this method in energy data application analysis compared to other methods. Is it necessary to use so many distance functions? Can they be classified? For example, the distance functions shown in Tables 9-10.

Comment 3) Visualization: Tables 5-7 do not highlight the key points and fail to reflect the advantages of distance functions. Figures 10-13 are not clear enough.

Comment 4) Format: P8 seems to be missing the heading "Section 3".

Comments on the Quality of English Language

 Moderate editing of English language required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Using Bayesian model selection, the research offers a unique solution to the cold-start issue in time series prediction. The deployment of an interactive visualization tool and the incorporation of previous knowledge are two powerful features that improve comprehension and application of the techniques covered. The approaches are firmly supported, and the literature evaluation is thorough.

-The use of Bayesian model selection is covered in the study, although it doesn't go into great detail on how it's done. More thorough explanations or examples might make the material easier to understand for people who are not familiar with Bayesian techniques. A detailed explanation of the Bayesian model selection process used in this specific situation would be beneficial.

 

-The incorporation of past information into the Bayesian model is discussed in the study. It does not, however, go into great depth on the type of previous knowledge that was applied, how it was chosen, or how it affected the model's functionality. Readers would comprehend the originality and usefulness of this technique better if there was more information available on these points.

-The quality of Figure 11, which visualizes the marginal likelihood, is very low.

-The suggested method's benefits are covered in the study, but a thorough comparison with other widely used time series prediction techniques is not given. The case for this method's adoption might be strengthened by providing comparison data or a description of how it performs in comparison to other approaches in terms of accuracy, computing efficiency, or applicability.

 

-The paper generally has good grammar and spelling, but there are a few issues:

--Sentence structure in Abstract: The sentence, "It instead applied statistical tests to validate model efficacy," could be clearer if rephrased as, "Instead, it applied statistical tests to validate model efficacy."

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

Accept after minor revisions like corrections to minor methodological errors and text editing.

Comments on the Quality of English Language

Minor editing of English language required.

Author Response

Comment: Accept after minor revisions like corrections to minor methodological errors and text editing.

Response: Thank you for your valuable feedback. We have incorporated the suggested revisions, with the edits highlighted in red. We appreciate your guidance throughout this process.

Reviewer 2 Report

Comments and Suggestions for Authors

The authors have addressed all of my concerns. I suggest accepting the paper

Author Response

Comment: The authors have addressed all of my concerns. I suggest accepting the paper.

Response: Thank you for your kind words and detailed review. We are pleased to hear that our revisions have addressed your concerns and we appreciate your recommendation for acceptance.