Detection of Emotion Using Multi-Block Deep Learning in a Self-Management Interview App

Round 1

Reviewer 1 Report

In this paper, the authors propose a deep learning approach for the detection of user emotions by using a Multi-Block CNN in a self-interview management system. Before the multi-block process, there is a phase for face detection, followed by an eyes and nose detection.

The major problem for this article is its readability. The problem does not derive from the English quality, style or grammar, but from the so many repetitions that distract from the most important aspects in each component of the system proposed. Furthermore, there are a lot of ambiguous  sentences.

Other aspects that can improve greatly the paper follow.

1. In the abstract, they talk about a “pretrained” AlexNet, but in the paper they never mentioned how the AlexNet was pretrained. Please can you clarify this aspect?

2. “For the mobile service configured in this study, an application is developed, utilizing Android Studio 5.1.1(Kitkat) on an Intel(R) Core(TM) i7-4770 CPU 3.40GHz, 16.00GB RAM-based Windows 8.1 Enterprise K 64-bit environment.” àThe configurations of Android and Windows are old. I would suggest to test more recent versions.

3. I suggest to test other state-of-the-art networks, in this way the Chapter 2 could be improved. 

4. Page 9, equations 2 and 3, it is necessary to give more information regarding the variables used in the equations.

5. Page 5 line 155. They mentioned the “CAS-PEAL Face database”. Please, better describe this dataset and also add the reference.

In conclusion, the problem addressed and the solutions proposed in this paper look interesting, but many major revisions are still required before accepting the paper for the publication.

Author Response

Dear Editor.

 

Please find our revised manuscript entitled “Detection of User Emotion using Multi-block Deep Learning in Self-Interview Management” submitted for consideration for publication in Applied Sciences.

 

 

Thanks for the reviewers through review of our manuscript, which greatly improved the quality of the manuscript.

 

On the separate sheets, the reviewers very helpful comments were addressed, point by point, in our response to each comment or revision request, with indication of pages on which the manuscript changes have been made.

 

We hope that after these corrections our manuscript will be acceptable for publication in Applied Sciences.

 

 

Sincerely yours.

Author Response File: Author Response.docx

Reviewer 2 Report

The paper deals with an interesting topic emotion analysis for supporting interview. In the present form the paper presents many lacks. Starting from the introduction authors should provide to the reader a wider picture on the research domain, introducing also some issues regarding the study done on 3D not only on 2D, also considering the improvements done both in the study done and the new technologies available also on mobile devices. Some more lines and references should be added as for instance the following ones:

VIOLANTE, Maria Grazia, et al. 3D Facial Expression Recognition for Defining Users’ Inner Requirements—An Emotional Design Case Study. Applied Sciences, 2019, 9.11: 2218.

Marcolin, et al. (2019, June). Emotional Design and Virtual Reality in Product Lifecycle Management (PLM). In International Conference on Sustainable Design and Manufacturing (pp. 177-187). Springer, Singapore.

Moos, S. et al (2010). Computer-aided morphological analysis for maxillo-facial diagnostic: a preliminary study. Journal of Plastic, Reconstructive & Aesthetic Surgery, 63(2), 218-226.

 

Also the methodological sections should be improved providing a more clear global overview of the methodology, also with the adoption of a graphical flowchart, for going further with the specific step that compose the method where it is necessary to provide more details reagrding the features selected for the methodology and whocha re the singeric elements that justify one paramters connected with the other adopted.

 

Also in the experimental validation it should be better to provide more detail regarding the experimental setting

 

Author Response

Dear Editor.

 

Please find our revised manuscript entitled “Detection of User Emotion using Multi-block Deep Learning in Self-Interview Management” submitted for consideration for publication in Applied Sciences.

 

 

Thanks for the reviewers through review of our manuscript, which greatly improved the quality of the manuscript.

 

On the separate sheets, the reviewers very helpful comments were addressed, point by point, in our response to each comment or revision request, with indication of pages on which the manuscript changes have been made.

 

We hope that after these corrections our manuscript will be acceptable for publication in Applied Sciences.

 

 

Sincerely yours.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

The authors address all the issues concerning my review. The paper can be accepted in the present form. 

Author Response

We thank the reviewer for the review of this manuscript, which greatly improved its quality. We have corrected the errors in technical writing, as well as the grammatical errors. The manuscript has been edited by a native English speaker.

At the time of the major revision, in order to correct English expressions according to the reviewers’ opinions, English editing was done through a native English speaker from “Nurisco co. ltd.”, an English proofreading company as the below contents. Nevertheless, if English editing is needed more, we will proceed with it again.

 

Author Response File: Author Response.docx

Reviewer 2 Report

The scientific level of the paper has been improved

Author Response

We thank the reviewer for the review of this manuscript, which greatly improved its quality. We have corrected the errors in technical writing, as well as the grammatical errors. The manuscript has been edited by a native English speaker.

At the time of the major revision, in order to correct English expressions according to the reviewers’ opinions, English editing was done through a native English speaker from “Nurisco co. ltd.”, an English proofreading company as the below contents. Nevertheless, if English editing is needed more, we will proceed with it again.

 

Author Response File: Author Response.docx