ISSA-ELM: A Network Security Situation Prediction Model
Yiming Ji
Round 1
Reviewer 1 Report
The authors proposed the ISSA-ELM method for network security situation prediction. ISSA optimizes the initial weights and ELM network for fast training. The proposed method is evaluated on 150 data collected from a network system. The overall quality of this manuscript is good.
1. Figure 6. Using a bar plot but not a line plot to visualize the data distribution is more appropriate.
2. Cross-validation is preferred for the small dataset with 150 samples.
3. It'd be better to briefly describe how to calculate the network security situation value.
4. ISSA is thoroughly evaluated by different benchmarks. However, the necessity of ELM for the network security application is not evaluated. How's the performance of other models, such as linear regression/SVR?
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This research proposed an ISSA-ELM network security prediction model, the research provided a good introduction to the involved algorithms and the model process. the research also provided a thorough study of the performance of the proposed model, using benchmark functions and a network experiment. To improve, the authors may want to update the abstract so that the contribution could be clear, simple and shorter sentences are preferred. The acronym may be introduced after the full phrase. In addition, Chinese characters may need to be replaced in figure 2; font size should be consistent in Figures 3 and 4 (font in figure 4 could be smaller); math equations in the text could also use a formula editor to edit them (in Section 5.1.5), and finally, the format of reference items should be consistent (item 18-20, 22-24, and 28-30).
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 3 Report
The paper is generally well written and based on an actual bibliography.
The only suggestion I have is that the authors should explain more in the text their graphs and results.
Also, at the end, the algorithms the results are compared with (namely ELM, GA-ELM, and SSA-ELM) should be referred accordingly.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
