Adapting Fleming-Type Learning Style Classifications to Deaf Student Behavior
Round 1
Reviewer 1 Report
The manuscript deals with the application of the machine learning approach for adapting Fleming-type Learning Style to Deaf Student Behavior.
My observations are
- The authors have provided a fair introduction to the proposed method along with its comparison with other methods by discussing their advantages and limitations. The motivation behind the machine learning approach is positively mentioned.
- An inclusive and up-to-date technical literature is presented to establish the importance of the proposed research, and the objective is clear.
- Section 2 helps readers to understand the algorithm used for data analysis, learning factors, learning style, learning for the deaf.
- The methodology has been illustrated with flowcharts that assist readers to understand the overall framework.
- The proposed 5-fold and 10-fold validations method is tested for assessing the effectiveness and found appropriate.
- Overall, a systematic analysis has been carried out that makes the presented research of good quality.
My suggestions
- Rephrase the sentence “This is the world’s first study to develop a novel integrated data fusion and assimilation technique to classify learning experiences and patterns among deaf students by employing Fleming’s with applied Thai Sign Language” as “This study presents the development of a novel integrated data fusion and assimilation technique to classify learning experiences and patterns among deaf students by employing Fleming’s with applied Thai Sign Language”.
- Authors must comment on the deployment of the model in real-time.
- The comparative study of various algorithms is provided however a robustness analysis of at least one algorithm (classifier) must be presented considering hyperparameter tuning/optimization. Refer following articles to lure potential readers:
- A White-Box SVM Framework and its Swarm-Based Optimization for Supervision of Toothed Milling Cutter through Characterization of Spindle Vibrations, preprint, arXiv, DOI: https://arxiv.org/abs/2112.08421
- A white-box model can be provided for understanding the classification of at least one classifier. Interpreting SVM or decision tree is easier. Refer following articles to lure potential readers and to develop a white-box model:
- A White-Box SVM Framework and its Swarm-Based Optimization for Supervision of Toothed Milling Cutter through Characterization of Spindle Vibrations, preprint, arXiv, DOI: https://arxiv.org/abs/2112.08421
- What about classification using a blind dataset (No labels)? Classification results must be provided considering the blind dataset (No labels). Apply the trained model to classify blind datasets. Refer to “Figure 8: Framework for classification of blind data” from the following article.
- Application of Bayesian Family Classifiers for Cutting Tool Inserts Health Monitoring on CNC Milling, International Journal of Prognostics and Health Management, Vol. 11 No. 2 (2020) https://doi.org/10.36001/ijphm.2020.v11i2.2929
- Justify why are machine learning algorithms preferred over deep learning ones? Refer to the following articles.
- Deep Learning Algorithms for Tool Condition Monitoring in Milling: A Review DOI: https://doi.org/10.1088/1742-6596/1969/1/012039
- There are a few spelling mistakes, and grammar is improvable in a few places. There must be thorough proofreading of the paper.
This investigation supports the ongoing revolution of academics. The presentation of the problem at hand is clear. The flow and continuity are appropriately maintained. Thus I recommend that the manuscript be accepted for publication contingent revisions as per my suggestions.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
Dear colleague,
I think your work is very interesting and can be of great help to improve the education of deaf students, nevertheless there are some aspects that could be improve.
The data analysis is correct and it is done in an intensive way, however there is something in the manuscript that I think could be enhance.
There are some mistakes in the wording could be corrected, for example, in the introduction a complete sentence is repeated: Once mined, the data are classified into learning style type that improve learning method, so it must be delated in one of the two repetitions.
Also I think that is importat to indicate the average age of the students, because in other parts of the world the academic levels are others, so it could be dificult to suppose how old have the students (it could be include in the abstract or in methodology).
In section 3.6.1 a questionnaire is prepared but at no time is a reliability index defined for it, an aspect that is done in section 3.6.2, so it seems necessary to include this data in the work.
And finally, in the discussion section, the use of any citation is not observed, this section must be supported by other works to be able to discuss the results, I think it should be rewritten using references from previous works.
Best regards.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
good paper. Impressive on my side indeed
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 4 Report
The description of research group (sample) is required.
It is not very clearly described: design research, hypotheses, questions, research period, context.
It is important to set research limits, especially in the case of a very small sample and the challenges of the research caried out to ensure the objectivity of the results obtained.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
They tried addressing my comments. Thus I recommend acceptance to publish. Congratulations.
Reviewer 4 Report
Great job!
