EmSM: Ensemble Mixed Sampling Method for Classifying Imbalanced Intrusion Detection Data

Round 1

Reviewer 1 Report

Please consider the following advice:

• Modify structure in the abstract and the rest of the document to improve readability by separating individual clauses instead of monolithic sentences.
• Typo line 38: remain/s
• Line 42: "exceed 1,000, which is considered to be immense" - According to whom?
• Several citations are missing references and/or cited in different styles.
• What was the feature selection methodology?
• Table 5 is redundant. Readers should already understand the concept of a confusion matrix. It is also a cut and paste from your previous work [27] and the article referenced in that paper also.
• Was the one-class classification problem considered?
• The conclusion is exceptionally brief, literally only two sentences. Please expand to bookend the problem you are addressing and your aims from the introduction.  

Author Response

It is an honor to have the opportunity to contribute our paper entitled, “EmSM: Ensemble Mixed Sampling Method for Classifying Imbalanced Intrusion Detection Data " to Electronics.

We appreciate your valuable comments. As requested, we have addressed reviewer’s comments, and have presented the corrections according to each point raised. 

Please see the attachment.

As the authors of this paper, we would like to express our gratitude for the theoretically and academically challenging comments that have been made, and for such a thorough review of our paper. The reviewers’ suggestions were extremely valuable. We hope that the revised manuscript satisfies your requirements and thank you again for your kind consideration.

Author Response File: Author Response.pdf

Reviewer 2 Report

The proposed work is a mixed resampling method using a hybrid synthetic minority oversampling technique with an edited neural network that increases the minority class and removes noisy data to generate a balanced dataset. The Paper is presented very well. Experimental results show the proposed method is very promising. Here are some comments:

• More information is needed for extracted features and methods applied in feature extraction.
• Why dataset tested by binary and multi-classification? What is the difference and importance to show in experimental results?
• Why different types of metrics are used? There are some other metrics n machine learning. What is the reason for the selection of these metrics?
• Minor typos they should be fixed.

Author Response

It is an honor to have the opportunity to contribute our paper entitled, “EmSM: Ensemble Mixed Sampling Method for Classifying Imbalanced Intrusion Detection Data " to Electronics.

We appreciate your valuable comments. As requested, we have addressed reviewer’s comments, and have presented the corrections according to each point raised. 

Please see the attachment.

As the authors of this paper, we would like to express our gratitude for the theoretically and academically challenging comments that have been made, and for such a thorough review of our paper. The reviewers’ suggestions were extremely valuable. We hope that the revised manuscript satisfies your requirements and thank you again for your kind consideration.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Thank you for your responses and for taking the time to consider and address them. Congratulations on your work, and I look forward to seeing more in the future.