A Novel Method for Cross-Modal Collaborative Analysis and Evaluation in the Intelligence Era
Round 1
Reviewer 1 Report
1. The contribution is not stated clearly.
2. The choice of parameters used in the algorithm is not well justified.
3. A deep and detailed comparison with other methods is mandatory.
4. The authors claim that their method is faster and more efficient, but this is not rigorously demonstrated since it is applied just for a particular case.
5. What do you mean by experimental validation? Where the data exactly comes from, what is their reliability and accuracy for which model? Please address this important point seriously. Authors must cite the following papers;
Overall the quality of this paper is very good.
I recommend this paper. Authors must cite the following papers:
Alrashed, Fahad Abdulaziz, Abdulrahman M. Alsubiheen, Hessah Alshammari, Sarah Ismail Mazi, Sara Abou Al-Saud, Samha Alayoubi, Shaji John Kachanathu et al. "Stress, Anxiety, and Depression in Pre-Clinical Medical Students: Prevalence and Association with Sleep Disorders." Sustainability 14, no. 18 (2022): 11320.
Ahmad, F., Shahid, M., Alam, M., Ashraf, Z., Sajid, M., Kotecha, K., & Dhiman, G. (2022). Levelized Multiple Workflow Allocation Strategy under Precedence Constraints with Task Merging in IaaS Cloud Environment. IEEE Access.
Singamaneni, K. K., Dhiman, G., Juneja, S., Muhammad, G., AlQahtani, S. A., & Zaki, J. (2022). A Novel QKD Approach to Enhance IIOT Privacy and Computational Knacks. Sensors, 22(18), 6741.
Singh, Shailendra Pratap, Wattana Viriyasitavat, Sapna Juneja, Hani Alshahrani, Asadullah Shaikh, Gaurav Dhiman, Aman Singh, and Amandeep Kaur. "Dual adaption based evolutionary algorithm for optimized the smart healthcare communication service of the Internet of Things in smart city." Physical Communication (2022): 101893.
Singh, S. P., Dhiman, G., Viriyasitavat, W., & Kautish, S. (2022). A Novel Multi-objective Optimization based Evolutionary algorithm for Optimize the Services of Internet of Everything. IEEE Access.
6. The authors should provide other applications of the proposed algorithm.
7. To demonstrate the effectiveness of the proposed algorithm a real experimental validation is mandatory for a rigorous and accurate comparison and validation.
8. Please discuss the performance of the technique for real-time applications?
9. Specify the limitations and drawbacks of the proposed method.
10. The conclusion must be rewritten.
Author Response
- Comment: The contribution is not stated clearly.
Response: Your question is very good. The contribution of this paper is proposing the method to improve the quality of emotional information obtained from student during the whole process of learning and make accurate evaluations of students' learning process. The relevant information is in section 1.2.
- Comment: The choice of parameters used in the algorithm is not well justified.
Response: Since we did not express the choice of parameters clearly, we are sorry for your misunderstanding. We have modified the corresponding parts in the Section 4.1 of the manuscript to make the expression more accurate and clearer.
- Comment: A deep and detailed comparison with other methods is mandatory.
Response: Your question is very reasonable. We have been actively looking for answers. For example, we have tried use comparative experiment in Section 4.3 to show the differences between all methods.
- Comment: The authors claim that their method is faster and more efficient, but this is not rigorously demonstrated since it is applied just for a particular case.
Response: Your suggestion is good. We should use more specific case to demonstrate the good performance of our method. But we don’t have enough time to collect the students' real states for the experiment. At the same time, we feel that the scope of the current thesis can still support the argument of this article. Therefore, we recommend that the supplementary experiment be included in another follow-up paper in the future.
- Comment: What do you mean by experimental validation? Where the data exactly comes from, what is their reliability and accuracy for which model? Please address this important point seriously. Authors must cite the following papers
Response: We are sorry for your misunderstanding about experimental validation. We have modified the corresponding parts in the Section 4 of the manuscript.
- Comment: The authors should provide other applications of the proposed algorithm.
Response: Thanks for your suggestion, our method is provided a method for multimodal sentiment analysis in STEAM education. In this paper we are only focus on an application, and we are going to continue to explore more applications in future follow-up papers.
- Comment: To demonstrate the effectiveness of the proposed algorithm a real experimental validation is mandatory for a rigorous and accurate comparison and validation.
Response: Thanks for your comment, we have discussed the experimental results in Section 4.4.
- Comment: Please discuss the performance of the technique for real-time applications?
Response: We have carefully evaluated the funding and experimental conditions required to complete this supplementary work and feel that we cannot afford this expanded scope of supplementary research at this time. At the same time, we feel that the scope of the current thesis can still support the argument of this article. Therefore, we recommend that the supplementary experiment be included in another follow-up paper in the future.
- Comment: Specify the limitations and drawbacks of the proposed method.
Response: We have added the limitations and drawbacks in Section 4.4.
- Comment: The conclusion must be rewritten.
Response: Thanks for your suggestion about conclusion, we have rewritten it.
Reviewer 2 Report
Paper should be corrected in:
1. English Grammar,
1.a. example: "In order to accurately captures emotional changes of learners and make an accurate evaluation" ... should be "In order to accurately capture emotional changes of learners and make an accurate evaluation"
1.b. example " Its main framework" should be "its' main framework"
2. Writing style:
2. a) Authors should be careful not to personalize the text. Example "is a key part of the development of my country's education" should be written without using "my" but neutrally...
2. b) Authors should avoid popular words in this text, but should make it be more moderate. Example of innapropriate use of words "deep-water area", "hot topic", "baton" and "steering wheel" ...
2. c) Authors should avoid specifics of particular countries and the inclusion of politics, but to focus on the essence of the manuscript contribution and make the contribution general. Example what is to be deleted and avoided in text: "Communist Party of China and the State Council issued"...
3. The use of abbreviations - example: first occurence of STEAM is in abstract, but any first occurence of an abbreviation should be followed or preceeded by the full name that was the basis for the acronym. In the whole text and tables, there are many abbreviations unexplained.
3. Content
3.1. Introduction is expected to provide a brief review of related work, at least at one paragraph and to contrast the findings with the contribution. Currently, there are no related work in the introduction. Too many specifics are written in introduction, related to China. This is scientific paper with general title - no case study performed in China was mentioned in title.
3.2. Introduction subtitle 1.1. Evaluation of STEAM education is not appropriate subtitle for this section, since the content does not evaluate the STEAM concept - there are too many precise sentences presenting the China education system reform.
3.3. Section 4. could not be entitled "Discussion" since it represents the Research methodology and results.
3.4. The whole paper emphasizes the emotions evaluation ("students' emotional state analysis during the learning process"), while the aim of this paper is to evaluate cross-modal collaboration in education process, based on STEAM education. It seems that this paper content is not aligned with the title. Example - "Moreover, during the process of students' learning, it can accurately grasp the emotional state of students by capturing the information of multiple modalities, adopting multi-modal emotional analysis and presenting evaluation results which could provide emotional guidance to students in a timely manner or adjust the difficulty level and teaching mode of the course to help students acquire knowledge much better and achieve the learning aims much easier."
3.5. The title of the manuscript is not precise "A novel method for cross-modal collaborative analysis and evaluation based on STEAM education" - what is the contribution of this paper ? This is not addressed in the title.
3.6. At Figure 1, there is a step in the proposed algorithm named "Concatenate and Softmax" which is totally unclear. In 3.2. Model architecture
3.7. In section 3.1. Problem definition, it is not explained enough why the emotional state is emphasized in the process of multimodal learning with STEAM approach.
3.8. STEAM education approach is not addressed in this paper properly, since the emphasis is on mix of media - text, video and audio material. It is expected to have different teaching subjects in curriculum combined, to have similar topics at different subjects parallely presented to students, to have the content of this paper aligned with the title. Currently, the title and the content are not aligned.
3.9. After introduction or within the introduction, there should be a background section to explain the terms and methods used in this approach.
3.10. The paper emphasizes the deep learning and artificial intelligence use in the evaluation of cross-modal collaborative analysis, but the proposed approach and research methodology is totally unclear - what is the proposed method of using deep learning (training, evaluation with real data upon the trained system) - how does this deep learning approach fit in the proposed model at Figure 1? What is the population used for training and evaluation...
3.11. The paper addresses the learning with students, but the data sets were not taken from students population, but from public CMU-MOSEI dataset, related to multimodal sentiment analysis.
3.12. Conclusion should be provided with several paragraphs - aims, results, contributions, limitations, future works. Currently, conclussion is messy and does not address what was promised in the title.
GENERAL CONTENT CONCLUSION: This paper seems to be a transformed version of some other paper that deals with sentiment analysis, not with evaluation of effects on cross-modal collaboration of teaching contents in STEAM education.
4. Technical correctness
Most images are blurred or too small to read.
Author Response
- Comment: English Grammar
Response: Thank you very much for discovering this error. We apologize for grammatical problems and have corrected it.
- Comment: Writing style
Response: Thank you very much for discovering this error. We apologize for writing style problems and have corrected it. Revised portions are marked in red in the paper.
- Comment: Content
Your suggestion is very good. We have modified the parts of the manuscript and changed some subtitle to make the expression more accurate and clearer. We have rewritten the concept of STEAM education in Line 44. And we have added the explanation about why we use proposed method to solve the problem in STEAM education.
- Comment: Technical correctness
Response: Thanks for your suggestion, we have already changed the image.
Reviewer 3 Report
1. The English in this paper seems like it was translated by a translator. English writing should be completely checked and revised, otherwise, this paper is hard to follow.
2. Line 82, what is "EGGs", the author should provide its full description.
3. Authors should provide the data source, e.g., video, text, and audio, which are utilized to train the learning model.
4. The font size in the image should be consistent with the font size of the text. For example, the font (i.e., Data alignment and normalization) in Figure 2 is too large.
5. Authors should specify the meaning of h_ij in Eq (2).
6. Why does the Chinese text appear in line 250 of this English academic paper?
Author Response
- Comment: The English in this paper seems like it was translated by a translator. English writing should be completely checked and revised, otherwise, this paper is hard to follow.
Response: Thank you very much for discovering this error. We apologize for grammatical problems and have corrected it.
- Comment: Line 82, what is "EGGs", the author should provide its full description.
Response: Thanks for discovering this error. We have provide its full description,
- Comment: Authors should provide the data source, e.g., video, text, and audio, which are utilized to train the learning model.
Response: Your question is very reasonable. We have provided the experimental data in Section 4.1.
- Comment: The font size in the image should be consistent with the font size of the text. For example, the font (i.e., Data alignment and normalization) in Figure 2 is too large.
Response: Thanks for your suggestion, we have already changed the image.
- Comment: Authors should specify the meaning of h_ij in Eq (2).
Response: Thanks for your suggestion, we have added the meaning of h_ij in Section 3.2.2.1.
- Comment: Why does the Chinese text appear in line 250 of this English academic paper?
Response: Thanks for your suggestion, we have deleted all improper text in our manuscript.
Reviewer 4 Report
Dear authors,
thank you for an interesting research.
Is is entitled:
- A Novel Method for Cross-Modal Collaborative Analysis and 2 Evaluation in the Intelligence Era
or
- A Novel Method for Cross-Modal Collaborative Analysis and 2 Evaluation Based on STEAM Education: Take Sentiment Anal- 3 ysis as an Example
?
In both cases, the attempt to measure the emotional envolvement is to be appreciated for the purposes to improve education' efficiency.
There are several questions:
- text on the lines 307-309 describes the sample, does the age is similar? E.g., the sentiments and emotions of 12-years old, 25-y.old and 50-y. old people would differ, wouldn't them?
- text on the lines 231-235 (and further, 253-254) clearly describes the modalities of text, audio, visual. Why the video is represented in the schemas separately? Do you distinguish specific features of video? In fact, you should distinguish them - the rythm, dynamic, move characteristics could be also interesting for the investigation of the video, and it differs from just a static visual data. Please, give some details about the difference between visual and video.
There is a typo:
- line 234 - the LSTM-Attienon (LSTM-Attention, I guess?)
Good luck to continue and to publish the final version!
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Round 2
Reviewer 2 Report
Paper should be corrected in:
1. title - Method for cross modal collaboration and evaluation of STEAM education - STEAM is abbreviation and should be ALL CAPS
2. title - does not reflect the content, since there is no emmotional aspect emphasized
3. Content - the particular mentioning of "Central commitee of the Communist Party of China" should not be mentioned, since this paper should have broader, international, non-political influence and impact. This issue has already been said in previous version of review.
3. Content - emphasis on emotional status during learning is not clearly enough related to cross modal ambiend of STEAM application, but it has been put generally.
4. Content - problem definition emphasize different learning media (visua, text, audio), but not multi-modal, i.e. cross modal analysis (which should be related to different contents from science, technology...according to STEAM abbreviation). This leads to conclusion that title is not aligned with the content.
5. Writing style - abbreviations at many spots, such as "BAM-model", "BLBA-MODEL" are not clear, why is it named like that?
6. Content - Softmax is not explained earlier in text, but the first occurence is in figure 1.
7. Content - The cross-modal analysis of STEAM education topic requires appropriate research dataset related to different contents from science, technology and their mutual impact, not only different media integration...while the dataset used in this research is used for general publicly available CMU-MOSEI dataset which is related to general sentiment analysis.
8. English writing - still have occassional errors, such as when a sentence starts with "And"...example: "And the method is applied to the data set..."
9. Conclussion - does not provide results explanation aligned with the title. not detained enough in future work part.
FINAL CONCLUSION - Generally speaking - this paper emphasizes different media (audio, video, text) and evaluation of sentiments based on general publicly available CMU-MOSEI dataset which is related to general sentiment analysis of persons (not necessarily students). Even these topics could be related to STEAM education, their connection to STEAM cross-modal analysis is not supported in the content of this paper. The paper content does not refer to the title. The paper could be accepted only if all previous issues are resolved (mentioned in this review and previous - since some required changes in review requests were not included in this version). Particularly, the most important issue is to be resolved- to align the title with the content of this paper - to have in title included: emotional impact of persons being exposed to different media (text, video, audio). Currently, the title and content of the paper do not align.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
I have no further concerns.
Author Response
Thank you! I have asked my English major colleagues to help me revise my English writing.
Round 3
Reviewer 2 Report
REQUIRED CHANGE FROM REVIEW 2: title - does not reflect the content, since there is no emotional aspect emphasized
TITLE SHOULD HAVE THE EMOTIONAL ASPECT EMPHASIZED, SINCE IT IS IN THE FOCUS OF THIS PAPER
REQUIRED CHANGE FROM REVIEW 2: Content - emphasis on emotional status during learning is not clearly enough related to cross modal ambient of STEAM application, but it has been put generally.
IN SECTION 2 IT HAS BEEN ANNOUNCED THAT THE CONTRIBUTION WILL BE RELATED TO “cross- modal collaborative analysis scheme for learning behavior is constructed, a more accurate cross-modal collaborative analysis model is established, and the robustness of the analytical model is improved using big data technology and pre-training.” THERE IS NO EVIDENCE THAT LEARNING BEHAVIOR WAS EXPERIMENTALLY EVALUATED, NOR THERE IS EVIDENCE OF USING BIG DATA TECHNOLOGY AND PRE-TRAINING WITH THE MACHINE LEARNING APPLICATION. THERE SHOULD BE DETAILED EXPLANATION ABOUT THE USE OF BIG DATA TECHNOLOGY AND MACHINE LEARNING APPLICATION. AUTHORS SHOULD BE VERY CAREFUL IN ANNOUNCEMENTS AND PROMISES, SINCE THEY SHOULD SUPPORT THEM IN THE CONTENT OF THE PAPER.
REQUIRED CHANGE FROM REVIEW 2: Content - problem definition emphasize different learning media (visua, text, audio), but not multi-modal, i.e. cross modal analysis (which should be related to different contents from science, technology...according to STEAM abbreviation). This leads to conclusion that title is not aligned with the content.
THE TERM “MULTI MODAL” HAS DIFFERENT MEANING IN STEAM EDUCATION. IT SHOULD BE RELATED TO DIFFERENT CONTENT FROM DIFFERENT SUBJECTS, SINCE STEAM EDUCATION INTEGRATES SCIENCE, TECHNOLOGY ENGINEERING, ARTS AND MATHEMATICS. IN THIS PAPER, MULTIMODAL APPROACH HAS BEEN FOCUSED ON COMBINING DIFFERENT MEDIA (TEXT, VIDEO, AUDIO ETC.), NOT NECESSARILY FROM DIFFERENT SUBJECTS = MATH, ART ETC. SO THE IDEA OF HAVING MULTIMODAL AND STEAM IN TITLE IS NOT SUPPORTED BY CONTENT.
“STEAM education is a holistic, interdisciplinary approach to learning that combines science, technology, engineering, arts and mathematics. It harnesses the natural symbiosis between these disciplines to foster creative problem-solving, collaboration and critical thinking.”
https://cie.spacefoundation.org/what-is-steam-education-and-why-is-it-important/
“empowering children to be both creative and critical thinking humans, engaged and active citizens, who can collaborate and work in diverse teams”. https://www.steam-ed.ie/
IN 1.2. IT HAS BEEN ANNOUNCED “this research collects students’ emotional status through the whole process of learning and conducts the collaborative analysis of learning behavior and thus constructs a new evaluation method of teaching which links learning behavior with learners' emotional status” . THIS IS NOT SUPPORTED BY RESULTS, SINCE THE DATA COLLECTION HAS NOT BEEN MADE WITH STUDENTS DATA, BUT FROM GENERALLY AVAILABLE cmu-mosei DATASET, WHICH HAS BEEN “The dataset contains more than 23,500 sentence utterance videos from more than 1000 online YouTube speakers.” http://multicomp.cs.cmu.edu/resources/cmu-mosei-dataset/
REQUIRED CHANGE FROM REVIEW 2: Writing style - abbreviations at many spots, such as "BAM-model", "BLBA-MODEL" are not clear, why is it named like that?
STILL THERE ARE ABBREVIATIONS NOT EXPLAINED BEFORE THE FIRST OCCURRENCE, SUCH AS “Bi-LSTM”. AUTHORS SHOULD VERY SERIOUSLY APPROACH TO THIS ISSUE. IT IS NOT POSSIBLE TO HAVE ABBREVIATION NOT EXPLAINED AT THE FIRST OCCURRENCE.
REQUIRED CHANGE FROM REVIEW 2: Content - The cross-modal analysis of STEAM education topic requires appropriate research dataset related to different contents from science, technology and their mutual impact, not only different media integration...while the dataset used in this research is used for general publicly available CMU-MOSEI dataset which is related to general sentiment analysis.
EMOTIONAL ASPECT IS NOT IN THE FOCUS OF STEAM EDUCATION, BUT CRITICAL AND CREATIVE THINKING. NEVERTHELESS, IT COULD BE ANALYZED, BUT AUTHORS SHOULD BE VERY CAREFUL IN TITLE FORMULATION – TO BE ALIGNED WITH CONTENT AND OF STATING THAT THEY COLLECTED DATA FROM STUDENTS’ LEARNING, WHILE THEY USE THE PUBLICLY AVAILABLE DATASET CREATED FROM YOUTUBE, WHICH IS NOT NECESSARILY RELATED TO STUDENTS.
REQUIRED CHANGE FROM REVIEW 2: Conclusion - does not provide results explanation aligned with the title. not detained enough in future work part.
CONCLUSION SHOULD HAVE COMMON STRUCTURE OF – WHAT WAS THE GOAL, WHAT WAS ACHIEVED, WHAT IS STILL OPENED, WHAT ARE FUTURE RESEARCH PLANS. CURRENTLY IT IS MESSY.
THE STATEMENT IN CONCLUSION IS NOT TRUE “Multi- modal learning analysis is an essential direction in the field of artificial intelligence…” WHILE THE REST OF THIS SENTENCE COULD BE ACCEPTABLE.
REQUIRED CHANGE FROM REVIEW 2: Particularly, the most important issue is to be resolved- to align the title with the content of this paper - to have in title included: emotional impact of persons being exposed to different media (text, video, audio). Currently, the title and content of the paper do not align.
THE MAYOR ISSUE HAS STILL NOT BEEN RESOLVED – THE TITLE OF THE PAPER REMAINS UNCHANGED AND STILL DOES NOT REFLECT THE CONTRIBUTION OF THE PAPER. THE PAPER HAS ESSENTIAL TERM “MULTIMODAL” MISUNDERSTOOD AND AS SUCH, NOT ALIGNED WITH THE INTERPRETATION OF THIS TERM IN STEAM EDUCATION CONTEXT. THE PAPER IS NOT DIRECTLY ALIGNED WITH STEAM EDUCATION, BUT MORE PRECISELY TO THE COMBINATION OF EDUCATIONAL MEDIA.
Author Response
Thank you very much for your comments and suggestions. Those comments are all valuable and very helpful for revising and improving our paper, as well as the important guiding significance to our research. We have revised the title, the conclusion, the explanations of each abbreviation and the parts that were not properly stated.
This research is a cross modal analysis under the STEAM education environment, and takes emotional analysis as an example. We define cross modal analysis that is analysis data comes from different modes (visual, text, audio), but not different contents from science, technology... At the same time, this study uses dataset to verify the effectiveness of research methods. We should use dataset about students’ behavior to demonstrate the good performance of our method. But we don’t have enough time to collect the students' real states for the experiment. At the same time, we feel that the scope of the current thesis can still support the argument of this article. Therefore, we recommend that the supplementary experiment be included in another follow-up paper in the future.
Round 4
Reviewer 2 Report
ENGLISH WRITING
There must be a better English writing checking. For example “and by optimize teaching methods” should be “and by optimizing…”…
SENTENCE COULD NOT START WITH “AND…”, example: “…behavior of each subject . And we will use different dataset related to the…”.
INTRODUCTION AND RELATED WORK
If the title remains to have STEAM, to explain more the concept of multimodality and cross-subject collaboration, to explain the use of big data or to remove it from the paper.
IN RESEARCH METHODOLOGY RESULTS THERE SHOULD BE EVIDENCE RELATED TO STATEMENTS.
IN 1.2. HAS BEEN SAID “Based on cross-modal learning analysis technology, collaborative analysis of student learning behavior data is carried out to mine and restore the students' learning emotional status information implied by the cross-modal and accompanying behavioral information in the classroom. A scientific model further reveals the implicit relationship between teaching and learning process information and classroom education quality and a new evaluation model of classroom teaching and learning has been constructed.”
THIS IS NOT SUPPORTED BY RESULTS, SINCE THE YOUTUBE VIDEOS WERE TAKEN, NOT STUDENTS LEARNING BEHAVIOR DATA, AS PROMISED
The paper promises “Taking the cross-modal learning behavior data as the carrier, with technologies such as artificial intelligence, big data etc., this research collects students’ emotional status through the whole process of learning and conducts the collaborative analysis of learning behavior and thus constructs a new evaluation method of teaching which links learning behavior with learners' emotional status. This research indicates significance in research and real-world practice in terms of achieving a comprehensive and accurate assessment of student's ability, cognitive level, personality traits, mental health, etc., explaining and exploring cross-modal data-driven educational phenomena and educational laws, by of- fering guidance on optimization of the teaching process,”
DATA WERE NOT COLLECTED DURING THE WHOLE PROCESS OF LEARNING, LEARNING BEHAVIOR WAS NOT ASSESSED, BUT YOUTUBE VIDEOS WERE USED.
EMOTIONS, BUT ATTENTION AND MEMORY – WHAT IS RELATION BETWEEN THEM
What types of emotions were the proposed methods related to, when only attention is emphasized. How are emotions related to attention?
“we proposed a BAM-model (Bimodal Attention Mechanism model), to capture the interaction information between modalities and contexts.“ “The aligned hid-den vectors are spliced as the input of the subsequent module; it’s the 222 LSTM-Attention (Long Short-Term Memory and Attention Mechanism) combination module. “ Later, taking the dataset from Youtube mentions: “Happiness, Sadness, Anger, Disgust, Surprise, and Fear.” These two aspects are not related in this paper – attention and all these: “Happiness, Sadness, Anger, Disgust, Surprise, and Fear.”
~~~
THERE IS NO EMOTION EXTRACTION (they should be extracted from educational process, when people exposed to educational media, i.e. combination of contents from STEAM group), CLASSIFICATION AND PREDICTION RESULTS OBVIOUSLY PRESENTED IN THIS PAPER. There is EMOTIONS CLASSIFICATION FROM TYPES OF VIDEOS FROM YOUTUBE- "Finally, through the fully connected layer and using the SoftMax function for emotion prediction and classification, the final original learned emotion and cognitive state of the model are obtained" WHAT IS LEARNED EMOTION? IN THE SENSE OF DEEP LEARNING AUTOMATION, THE NEURAL NETWORK WAS TRAINED TO SOME EMOTION RECOGNITION….THE CONCEPT OF LEARNED EMOTION SHOULD BE MORE EXPLAINED.
RESEARCH METHODOLOGY SHOULD BE EXPLAINED WITH MORE DETAILS
· Big data methods is promised, but there is no explanation of the use of big data methods. Only large data sets imply taking big data, but which methods were used. Authors should provide more detailed list and explanations of all methods used.
· It is not clear how the research was conducted at all – videos were taken from you tube, representing some emotions. But, were these persons at these videos exposed to different teaching material – video, text, audio…to judge on their emotional response…it is not clear…
PAPER TITLE - STEAM EDUCATION mentions in title require steam content in the paper or to delete STEAM from the title
Steam education is not clearly explained and related to diffeent media. MODULES FROM STEAM EDUCATION ARE RELATED BY CONTENT. THEY COULD BE THE SAME MEDIA (TEXTS related, IN ALL DIFFERENT SUBJECTS, i.e. contents) OR DIFFERENT MEDIA.
the term multimodal was not explained in the context of steam education, but it has been taken as granted, as if multimodal in steam is the multimedia in fact.
the context, i.e. content that is to be related between different media has been only mentioned, but it is the essence of the steam approach – to relate content of the subjects topics…this is not supported here. so mentioning steam in the title is not supported by research methodology and results and should be removed from the title or better supported with the paper content.
“STEAM EDUCATION IS A HOLISTIC, INTERDISCIPLINARY APPROACH TO LEARNING THAT COMBINES SCIENCE, TECHNOLOGY, ENGINEERING, ARTS AND MATHEMATICS. IT HARNESSES THE NATURAL SYMBIOSIS BETWEEN THESE DISCIPLINES TO FOSTER CREATIVE PROBLEM-SOLVING, COLLABORATION AND CRITICAL THINKING.”
HTTPS://CIE.SPACEFOUNDATION.ORG/WHAT-IS-STEAM-EDUCATION-AND-WHY-IS-IT-IMPORTANT/
“EMPOWERING CHILDREN TO BE BOTH CREATIVE AND CRITICAL THINKING HUMANS, ENGAGED AND ACTIVE CITIZENS, WHO CAN COLLABORATE AND WORK IN DIVERSE TEAMS”. HTTPS://WWW.STEAM-ED.IE/
CONCLUSION – NOT SUPPORTED BY RESULTS
“this paper proposes a multi-modal analysis method to obtain students’ sentiment status through the whole learning process and make accurate evaluations of students' learning process. All the above experiments demonstrate this study method’s effectiveness and excellent performance.”
THE WHOLE LEARNING PROCESS WAS NOT INCLUDED IN RESEARCH RESULTS. NO EVALUATION OF STUDENTS’ LEARNING HAS BEEN SUPPORTED, SINCE NO STUDENTS’ DATA WERE INCLUDED, BUT ONLY YOUTUBE VIDEOS.
CONCLUSION
the paper should emphasize the true content of the paper – people attention in exposing to different media
no steam education OR EDUCATION AT ALL, no emotions ARE TO BE MENTIONED IN TITLE…
or to significantly improve the paper to have emotions like anger etc. related to attention, to have the data set obtained from educational environment which has different topics from steam set of subjects.
Author Response
We are very grateful to you for reviewing the paper so carefully. We have carefully considered the suggestion and make some changes. We have revised the title, the abstract, the conclusion and the parts that were not properly stated, especially the contents about STEAM education. Now we have removed STEAM from the paper and replaced it with intelligent education. Since we did not express “Attention” clearly, we are sorry for your misunderstanding. The “Attention” does not mean notice taken of someone or something, the “Attention mechanism” is an important algorithm in deep learning model, which is used to allocate computing resources to more important tasks.
