Denoising Masked Autoencoder-Based Missing Imputation within Constrained Environments for Electric Load Data
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis paper introduces the denoising masked autoencoder (DMAE) model as a solution to improve the handling of missing data, even in such restrictive settings. The logic of the proposed control strategy is clear. Some results show the effectiveness of the proposed method. This paper is well-presented and easy to follow. There are some suggestions that should be considered to improve the manuscript.
(1) It is suggested that the difference between this paper and other papers should be clearly listed and stated in the manuscript, especially the novelty compared with the existing research, and the research gaps should be highlighted.
(2) The literature review part could be improved, and more related literature could be added since the research status of the introduction is insufficient.
(3) All the results and comparisons have proved the feasibility of the proposed method. However, it is suggested that the limitations and shortcomings of this proposed method should be mentioned in this paper. Future research prospects should be mentioned in the last.
Comments on the Quality of English LanguageTHere is no comments about language.
Author Response
We appreciate the reviewers’ comments and suggestions to improve the quality of this manuscript. We have tried our best to understand and answer all the comments provided. In the attached file, we present point-by-point discussions about the reviewers’ concerns, along with the changes that were made. To enhance readability, the comments are in italicized font while our discussion is in black font. In addition we also quote the revised or added contents in blue font for reviewer’s convenience. We hope that the discussions below and revisions address the substantive concerns of the reviewer. If we have overlooked an aspect that would improve the paper we are willing to do so.
Sincerely,
Jaeik Jeong
Tai-Yeon Ku
Wanki Park
Author Response File: Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsThe authors have presented a novel denoising masked autoencoder (DMAE) model to manage missing data in energy efficiency research. However, the manuscript could greatly benefit from a more comprehensive theoretical analysis to support the workings of this model.
The authors have clarified how the DMAE integrates the learning of correlations between observed and missing values using the denoising autoencoder (DAE), and auto-correlations between observed values using the masked autoencoder (MAE). Nevertheless, a more detailed explanation of the fundamental principles and mathematical models behind the successful operation of DMAE in handling missing values would significantly enhance the reader’s understanding.
Moreover, I suggest that the authors may find it beneficial to consider the book “Deep Learning” by Goodfellow, Bengio, and Courville. This resource provides an extensive overview of various principles and techniques in deep learning, which could add to the theoretical depth of the manuscript. It may offer additional insights into improving or interpreting the DMAE model's results.
Additionally, a comparative study with other existing models used in energy efficiency research could provide a more holistic view of the effectiveness of the DMAE model. This comparison could also highlight the unique advantages of the proposed model.
Lastly, the authors could also consider involving more diverse datasets in their simulation experiments to further validate the robustness and universality of the DMAE model.
Overall, these enhancements will significantly improve the rigor of the manuscript and make it a more substantial contribution to the field of energy efficiency research.
Comments on the Quality of English Languageok
Author Response
We appreciate the reviewers’ comments and suggestions to improve the quality of this manuscript. We have tried our best to understand and answer all the comments provided. In the attached file, we present point-by-point discussions about the reviewers’ concerns, along with the changes that were made. To enhance readability, the comments are in italicized font while our discussion is in black font. In addition we also quote the revised or added contents in blue font for reviewer’s convenience. We hope that the discussions below and revisions address the substantive concerns of the reviewer. If we have overlooked an aspect that would improve the paper we are willing to do so.
Sincerely,
Jaeik Jeong
Tai-Yeon Ku
Wanki Park
Author Response File: Author Response.pdf
Reviewer 3 Report
Comments and Suggestions for AuthorsIn my opinion, this paper is technically sound enough to publish in this journal with a minor edition. My comments are below:
The flow of this paper can be improved, especially the introduction (literature review portion), by adding a few more recent articles.
The proposed DMAE algorithm needs more description for a better explanation.
The abstract and conclusion section could be better.
N/A
Author Response
We appreciate the reviewers’ comments and suggestions to improve the quality of this manuscript. We have tried our best to understand and answer all the comments provided. In the attached file, we present point-by-point discussions about the reviewers’ concerns, along with the changes that were made. To enhance readability, the comments are in italicized font while our discussion is in black font. In addition we also quote the revised or added contents in blue font for reviewer’s convenience. We hope that the discussions below and revisions address the substantive concerns of the reviewer. If we have overlooked an aspect that would improve the paper we are willing to do so.
Sincerely,
Jaeik Jeong
Tai-Yeon Ku
Wanki Park
Author Response File: Author Response.pdf