FSCB-IDS: Feature Selection and Minority Class Balancing for Attacks Detection in VANETs

Round 1

Reviewer 1 Report

In this article, the author proposes an enhanced intrusion detection framework for Vehicular Ad Hoc Networks (VANETs). The framework utilizes mutual information to select the most relevant features for constructing an effective model and employs the Synthetic Minority Over-sampling Technique (SMOTE) to address class imbalance issues. The structure is comprehensive, and the ideas are well-defined. but there are some areas that need improvement to enhance the quality and readability of the article. Here are some specific suggestions:

1. The abstract is excessively verbose and needs to highlight the author's main contributions. Additionally, the author excessively uses the term 'we' in the abstract.

2. The referencing in your paper requires attention. References 1 and 49 are missing, and you should include them as appropriate. Additionally, the citation order and formatting throughout the paper lack consistency. I strongly recommend revising and standardizing the citation style, ensuring accurate and complete referencing.

3. The author fails to provide a summary of the previous techniques before the third section, which hinders the support for the author's research. Additionally, there is a lack of emphasis on the author's contributions throughout the article.

4. The author includes Figure 1 in the background, but it is not referenced in the background.

5. The author mentioned the sampling step in line 244; you should add a number to make it clearer.

6. In lines 249 and 169 of the paper, there is no explanation provided for the meaning of the formulas.

7. There are two periods at the end of the line 259 in the paper.

8. Can the content of Section 3.4, which consists of only three sentences, be considered as a separate section?

9. The quality of the data figures 3, 4, 5, 9, and 11 in the article is poor. The images are blurry, making it difficult to discern the axis labels clearly. Additionally, there is data overlap, making it impossible to read the data accurately.

10. The author's analysis of figures 3 and 4 in the paper is not sufficiently comprehensive. Please provide additional details for this section.

11. While you compare various machine learning techniques and evaluate them on three different datasets, the analysis of the simulation results is too brief and lacks diversity in presentation. I urge you to provide more detailed explanations and thorough analyses of the simulation results. Additionally, figures such as Figure 7 and 13 et al should be discussed in the analysis to provide a comprehensive evaluation of your work.

12. The quality of figures 8 and 14 in the paper is poor. The data lines are not clear, and it is recommended to zoom in on the specific areas.

13. The formatting of tables 3, 4, and 5 in the paper is incorrect and does not comply with the triple-line format.

14. In the conclusion section of the paper, the main contributions should be highlighted in detail. Additionally, the major limitations and their implications for future research should be described.

Minor editing of English language required.

Author Response

 

Comment 1: The abstract is excessively verbose and needs to highlight the author's main contributions. Additionally, the author excessively uses the term 'we' in the abstract.

Response:

Thank you for your careful review. We have revised the abstract, limited the use of the term "we," and expanded the section that presents our contributions. Additionally, we have included the experimental results that demonstrate the performance of our model.

Comment 2: The referencing in your paper requires attention. References 1 and 49 are missing, and you should include them as appropriate. Additionally, the citation order and formatting throughout the paper lack consistency. I strongly recommend revising and standardizing the citation style, ensuring accurate and complete referencing.

Response:

Thank you for bringing the referencing issues to our attention. We have reviewed your feedback and made the necessary revisions to include the missing references in their appropriate places.

Comment 3: The author fails to provide a summary of the previous techniques before the third section, which hinders the support for the author's research. Additionally, there is a lack of emphasis on the author's contributions throughout the article.

Response:

Thank you for your feedback. I have carefully considered your suggestion and have made the necessary revisions. I have added technical summaries for each of the references, providing a comprehensive overview of the existing techniques and their limitations. This addition helps establish the context and support for my research. Furthermore, I have incorporated a paragraph after the table that specifically highlights the limitations of the previous studies and underscores the advantages of my work. This addition emphasizes the novel aspects of my research and its contributions to the field.

Comment 4: The author includes Figure 1 in the background, but it is not referenced in the background.

Response:
Thank you for your comment. I have reviewed the manuscript and made the necessary updates to ensure that Figure 1 is appropriately referenced in the background section. The figure now complements the relevant information and enhances the understanding of the topic. Please let me know if there are any further adjustments or clarifications needed.

Comment 5: The author mentioned the sampling step in line 244; you should add a number to make it clearer.

Response:

Thank you for your feedback. We have revised the manuscript and included a clear reference to the sampling step for better clarity and understanding.

Comment 6: In lines 249 and 169 of the paper, there is no explanation provided for the meaning of the formulas.

Response:

Thank you for your valuable revision. I have revised the paper to include detailed explanations for the formulas in both lines. The meaning and significance of each formula are now clarified, providing a better understanding for the readers.

Comment 7: There are two periods at the end of the line 259 in the paper.

Response:

Thank you for bringing this to my attention. I have made the necessary correction to remove the extra period and ensure the proper punctuation.

Comment 8: Can the content of Section 3.4, which consists of only three sentences, be considered as a separate section?

Response:

Thank you for your feedback regarding the content of Section 3.4. Upon careful consideration, I have made the necessary adjustments based on your suggestion. I have combined the short subsection with the previous subsection (3.3) and renamed it "Data Normalization and Classification" to ensure a more cohesive and comprehensive section.

Comment 9: The quality of the data figures 3, 4, 5, 9, and 11 in the article is poor. The images are blurry, making it difficult to discern the axis labels clearly. Additionally, there is data overlap, making it impossible to read the data accurately.

Response:

Thank you for your careful review, I have made improvements to the quality of Figures 3, 4, 5, 9, and 11 in the article. I have addressed the issue of blurriness and enhanced the overall clarity of the images. The axis labels have been adjusted to ensure readability, and measures have been taken to eliminate data overlap, allowing for accurate data interpretation. These enhancements aim to improve the visual experience and facilitate a better understanding of the figures.

Comment 10: The author's analysis of figures 3 and 4 in the paper is not sufficiently comprehensive. Please provide additional details for this section.

Response:

Thank you for your feedback. I appreciate your comment regarding the analysis of Figures 3 and 4 in the paper. Upon careful consideration, I have made the necessary adjustments to provide a more comprehensive analysis of these figures.

Comment 11: While you compare various machine learning techniques and evaluate them on three different datasets, the analysis of the simulation results is too brief and lacks diversity in presentation. I urge you to provide more detailed explanations and thorough analyses of the simulation results. Additionally, figures such as Figure 7 and 13 et al should be discussed in the analysis to provide a comprehensive evaluation of your work.

Response:

Thank you for your feedback. We appreciate your suggestion to provide more detailed explanations and thorough analyses of the simulation results. In response, we have made significant improvements to the analysis section to address this concern. We have expanded upon the discussion of the simulation results, providing comprehensive evaluations of our work.

Furthermore, we have included a discussion of figures such as Figure 7 and 13, among others, in our analysis. These figures play a crucial role in presenting additional insights and supporting our findings.

Comment 12: The quality of figures 8 and 14 in the paper is poor. The data lines are not clear, and it is recommended to zoom in on the specific areas.

Response:

Thank you for your feedback. We appreciate your suggestion to improve the quality of figures 8 and 14 in the paper. We have made adjustments to the size of the figures to ensure that they are easily readable. However, we would like to note that the close proximity of the values on the curves may make it challenging to distinguish them visually. Nevertheless, we have provided labels on the side of the figures indicating the values for each machine learning technique. This allows readers to easily interpret the figures and differentiate between the different models. We believe that these adjustments will enhance the readability and understanding of the figures in the paper.

Comment 13: The formatting of tables 3, 4, and 5 in the paper is incorrect and does not comply with the triple-line format.

Response:

Thank you for bringing the formatting issue of tables 3, 4, and 5 to our attention. We have carefully reconstructed the tables, ensuring compliance with the triple-line format as per your suggestion.

Comment 14: In the conclusion section of the paper, the main contributions should be highlighted in detail. Additionally, the major limitations and their implications for future research should be described.

Response:

Thank you for your valuable feedback on the conclusion section of our paper. We appreciate your suggestion to highlight the main contributions in more detail and discuss the major limitations and their implications for future research. We have carefully revised the conclusion to address these points.

Reviewer 2 Report

This manuscript proposes an enhanced intrusion detection framework for vehicular adhoc networks (VANETs) that leverages mutual information to select the most relevant features for building an effective model and synthetic minority oversampling (SMOTE) to deal with the class imbalance problem. However, it has following issues to be addressed:

1. In the abstract, there is a need to discuss the constraints a bit more.

2. Some of the acronyms have been mentioned directly. I suggest to please define their full form first and later acronym may be used in that particular page.

3. Keywords may be written in alphabetical order. 

4. Literature review needs to be expanded a bit more please add more papers. Few of them are suggested below: 

a. Abro, Ghulam E. Mustafa, Saiful Azrin BM Zulkifli, Kundan Kumar, Najib El Ouanjli, Vijanth Sagayan Asirvadam, and Mahmoud A. Mossa. "Comprehensive Review of Recent Advancements in Battery Technology, Propulsion, Power Interfaces, and Vehicle Network Systems for Intelligent Autonomous and Connected Electric Vehicles." Energies 16, no. 6 (2023): 2925.

b.  Rehman Javed, Abdul, Zunera Jalil, Syed Atif Moqurrab, Sidra Abbas, and Xuan Liu. "Ensemble adaboost classifier for accurate and fast detection of botnet attacks in connected vehicles." Transactions on Emerging Telecommunications Technologies 33, no. 10 (2022): e4088.

c. Alhaidari, Fahd A., and Alia Mohammed Alrehan. "A simulation work for generating a novel dataset to detect distributed denial of service attacks on Vehicular Ad hoc NETwork systems." International Journal of Distributed Sensor Networks 17, no. 3 (2021): 15501477211000287.

5.  I believe the table 1 shows the state of the art approaches but please explain it a bit more in terms of constraints that they have and those which you are going to resolve. Be specific in this regard. 

6. Conclusion is too small and technical analysis on results is missing. Please add them in it.

 

Yes regarding English language of the paper. I have observed some minor grammatical and punctuation errors. Please rectify them with by re-reading the entire manuscript please. 

 

Author Response

1. Comment 1: In the abstract, there is a need to discuss the constraints a bit more.

Response:

Thank you for your valuable feedback. We appreciate your suggestion to provide a more detailed discussion on the constraints in the abstract. In light of this, we have revised the abstract to elaborate on the specific constraints faced by Vehicular Ad Hoc Networks (VANETs). We now highlight the challenges posed by the fast mobility of vehicles and dynamic changes in network typology, which introduce rapid variations in network topology and add complexity to maintaining stable and reliable communication. These constraints emphasize the need for robust and efficient solutions tailored to VANETs, ensuring secure and reliable communication in dynamic and fast-changing environments.

2. Comment 2: Some of the acronyms have been mentioned directly. I suggest to please define their full form first and later acronym may be used in that particular page.

Response:
Thank you for your feedback. Upon reviewing the paper, we have made the necessary revisions to define the full form of each acronym upon its first mention in the text.

3. Comment 3: Keywords may be written in alphabetical order. Literature review needs to be expanded a bit more please add more papers. Few of them are suggested below: 
4. Abro, Ghulam E. Mustafa, SaifulAzrin BM Zulkifli, Kundan Kumar, Najib El Ouanjli, VijanthSagayanAsirvadam, and Mahmoud A. Mossa. "Comprehensive Review of Recent Advancements in Battery Technology, Propulsion, Power Interfaces, and Vehicle Network Systems for Intelligent Autonomous and Connected Electric Vehicles." Energies16, no. 6 (2023): 2925.
5. RehmanJaved, Abdul, ZuneraJalil, Syed AtifMoqurrab, Sidra Abbas, and Xuan Liu. "Ensemble adaboost classifier for accurate and fast detection of botnet attacks in connected vehicles." Transactions on Emerging Telecommunications Technologies33, no. 10 (2022): e4088.
6. Alhaidari, Fahd A., and Alia Mohammed Alrehan. "A simulation work for generating a novel dataset to detect distributed denial of service attacks on Vehicular Ad hoc NETwork systems." International Journal of Distributed Sensor Networks17, no. 3 (2021): 15501477211000287.

Response:

Thank you for your feedback. We have taken note of your suggestions and made the necessary adjustments. The keywords have been arranged in alphabetical order for improved readability. Additionally, we have expanded the literature review by including the suggested papers to provide a more comprehensive overview of the research landscape in the field. We believe that these additions enhance the quality and depth of the study.

4. Comment 4: I believe the table 1 shows the state of the art approaches but please explain it a bit more in terms of constraints that they have and those which you are going to resolve. Be specific in this regard. 

Response:

Thank you for your valuable feedback. We have taken your suggestion into consideration and have thoroughly addressed the constraints encountered by the studies presented in Table 1. We have also emphasized how our proposed methodology effectively addresses these constraints by utilizing advanced techniques, including mutual information for feature selection and the SMOTE algorithm for handling class imbalance. By doing so, we have enhanced the scalability, adaptability, and overall performance of our intrusion detection system for Vehicular Ad Hoc Networks (VANETs).

5. Comment 5: Conclusion is too small and technical analysis on results is missing. Please add them in it.

Response:

Thank you for your feedback. We have taken your suggestion into consideration and have expanded the conclusion to include a more detailed technical analysis of the results. We have also addressed the constraints faced by the studies mentioned in Table 1 and explained how our approach resolves them.

 

6.  comment 6: Conclusion is too small and technical analysis on results is missing. Please add them in it.

Response : We have modified the conclusion and added requested information

 

Reviewer 3 Report

I think your study titled "Feature Selection and Minority Class Balancing for Attacks Detection in VANETS" is quite interesting. Please consider my comments below.

 

1. The Abstarct section should be revised in a more understandable way for readers.

2. More information on the subject should be given in the introduction.

3. The novelty, contribution and difference of the study should be clearly demonstrated.

4. Verbal expressions should be added to the Result&Discussion section.

5.Conclusion section should be revised.

Author Response

• Comment 1 : The Abstract section should be revised in a more understandable way for readers.

Response : Thank you for your valuable feedback. We appreciate your suggestion to provide a more detailed discussion on the constraints in the abstract. In light of this, we have revised the abstract to elaborate on the specific constraints faced by Vehicular Ad Hoc Networks (VANETs). We now highlight the challenges posed by the fast mobility of vehicles and dynamic changes in network typology, which introduce rapid variations in network topology and add complexity to maintaining stable and reliable communication. These constraints emphasize the need for robust and efficient solutions tailored to VANETs, ensuring secure and reliable communication in dynamic and fast-changing environments.

 

 

• Comment 2 : More information on the subject should be given in the introduction.

We would like to think your for this remark: please note that we think that introduction contain the sufficient information. Furthermore, the background and related work are described in separated section

• Comment 3: The novelty, contribution and difference of the study should be clearly demonstrated.

 

The main contribution of this paper is summarized in the abstract and conclusion. Furthermore, the Results and Discussions section describe the details of obtained results

• Comment 4: Verbal expressions should be added to the Result &  Discussion section.

It is done

• Conclusion section should be revised.

 

Response : We have modified the conclusion and added requested information

Round 2

Reviewer 1 Report

The improved version of your article is definitely better than the previous one. I did more than I expected to improve the article.

There is, however, a format problem with Tables 1-2.