Multi-Output Based Hybrid Integrated Models for Student Performance Prediction
Round 1
Reviewer 1 Report
Abbreviations should be defined the first time they appear in the text, for example, IDA-SVR
There are typos for example, instead of full stop comma is used. Also I have noticed that there are in some cases no space after full stop and beginning of next sentence, and no space after comma. Also, there are errors like next “In [6], The”.
In Multi-output Prediction Model Framework it is stated “The input data of the model framework includes student names, task clicks, video viewing time and other data.” This is a problem since the private student data is acquired. How is this data protected?
Figure 1. does not have adequate possession in text.
There is text “Proof of” which does not relate to anything.
Figure 3. Standard deviation of some data, “some” data is not adequate term.
Figure 4 is missing. If it is the one on the bottom of the 7th page it is with low quality.
Table 4 is confusing.
Figure 9. Compari, also confusing.
Formatting of manuscript is like that that is very difficult to read.
In Results section it is not stated time period for experiment.
Author Response
point 1：figure 1.does not have adequate possession in text
response 1：Sorry, I didn't understand the meaning of this paragraph. I optimized the framework part
point 2：Table 4 is confusing.
response 2：Table 2 is a commonly used confusion matrix table in machine learning, providing proof of the formula source of measurement indicators
point 3：Figure 9. Compari, also confusing.
response 3：Figure 9 compares the best performing model under Method 1 and the best performing model under Method 2 to find the best model. It also compares which method is the best between Method 1 and Method 2 and has been modified
Other modifications are submitted as documents. Thank you for your comments
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In this paper, authors used learning data from the students' first six weeks, of course, plat- 409 to predict the students' homework, experiment, mid-term and final results after six 410 weeks. The authors used 14 Xgboost models to predict the midterm and final grades, and the accuracy reaches 78.37%, which is 15 3%-8% higher than the comparison model. Again, the Gdbt model is used to predict the homework and experimental grades, and the mean square error average is 16.76, which is better than the comparison of 17 models. Results show prediction results are helpful and achieved good results.
The author should revise this manuscript on the following points:
1. The motivation of this paper could be clarified. It is important to show the reason for doing this research.
2. Improve the introduction part by making it more concise.
3. The author should include one more section “experimental setup” after methods and materials and explain the complete experimental setup.
4. The author should replace results with Experimental results and discussion in the section 4.
5. The author should give the parameter Settings of the proposed algorithm and the comparison algorithm.
6. The author should supplement the literature on 2023 and need to add more works that are relevant.
Author Response
Thank you for your comments. I have modified it and uploaded it as a file
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
Replaced chart (Figure 4. Multi-output prediction model) contains non-english words so it is not understandable. All text should be in english.
"3.7.1 Comparison of Classification Models" is difficult to read because the text is behind the table 6.
"Figure 9. Compari" authors did not resolved this problem.
Table 9. is above text, so it is not readeble.
There are two sections no 5. Maybe there is no need for using "5. Experimental results and discussion". The text can be under section 4.
Author Response
point 1:"Figure 9. Compari" authors did not resolved this problem.
response:"Compari" is changed Comparison
Author Response File: Author Response.pdf
