A Time-Series Feature-Extraction Methodology Based on Multiscale Overlapping Windows, Adaptive KDE, and Continuous Entropic and Information Functionals

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

All the terms in the equations must be described. e.g. equation 9.

The authors stated that "There are various methods available in the literature for inferring PDFs from a time series window, and no single method is considered the best."  Few references may be provided in this regard

The values obtained for Mean Integrated Squared Error (MISE) for the real data analysis may be mentioned.

The limitation of the proposed methodology may also be mentioned in the conclusion

Author Response

In the attached PDF, you will find the cover letter listing the main changes applied to the article, along with the point-by-point response to the reviewer

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The methods in related work should be compared in the experiment section.

The related work should be updated.

The introduction identifies the limitations of existing methods but lacks a comprehensive comparison with specific state-of-the-art approaches. A more detailed analysis, highlighting the advantages of the proposed method over existing techniques, would strengthen the paper's contribution.

What are the novelties? The authors should explain more.

The author did not cite deep learning methods in the related work section, and a review of deep learning methods is required.

Comments on the Quality of English Language

This manuscript is generally well-written, but could be further improved through minor revisions to enhance readability and academic tone. Some sentences are excessively long and complex, and could be simplified to enhance clarity and conciseness. 

Author Response

In the attached PDF, you will find the cover letter listing the main changes applied to the article, along with the point-by-point response to the reviewer

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

The authors presented a time series feature extraction methodology based on multiscale overlapping windows, adaptive KDE, and continuous entropic and information functionals. After careful reading, there are some issues which should be addressed by the authors:

1. An important issue of the considered model is to get faster convergence speed. Which factors will influence the convergence speed?

2. The authors proposed the non parametric equations, what happen when parametric equ are considered in (1),(2),(3).

3. What is \mathbb{R} in Eq (2) and (3)?

4. The introductions are too big and authors should deleted the unwanted essays, highlighted the contributions over other paper.

5. What’s the limitation of your proposed model? Are there other ways that the results can be further improved?  One or two remarks should be given to discuss it in detail in the revised paper. 

6. How to get the parameter values in Table 1 and 2, need some discussion.

7. What is the relation between the FTSN and fuzzy in the Theorems and Examples, where the fuzzy impact in your problem formulation. Authors should clearly mentioned in the paper.

8. What is the connection between Jensen Shannon Divergences, kernel method, and generalised 16 entropy, need comparison with the recent literatures.

Comments on the Quality of English Language

nil

Author Response

In the attached PDF, you will find the cover letter listing the main changes applied to the article, along with the point-by-point response to the reviewer

Author Response File: Author Response.pdf

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

The authors addressed all the comments.

Comments on the Quality of English Language

nil