Review Reports - A Mixed Clustering Approach for Real-Time Anomaly Detection

Round 1

Reviewer 1 Report

Summary and comments

The paper titled “A Mixed Clustering Approach for Real Time Anomaly Detection” put forwarded an anomaly detection algorithm for real time data with mixed attributes using clustering techniques. Further, the algorithm starts with partitioning approach then follows agglomerative hierarchical approach.

Well, this is an interesting study, and I have following major comments to improve the article further.

1. The proposed model for real time data is based on a synthetic data set. It would be useful if application of the method is done on a real-world data set by creating a separate section (Assessing the performance, (please see: https://www.nature.com/articles/s41598-021-83340-8) and mentioning the results and model accuracy after that.

2. The study is conducted on a synthetic dataset (Line 459). Perhaps, a paragraph on limitations of the study must be included in the conclusion section, or create another section of limitations of the study, whichever is suitable.

3. For comparative studies only two methods are chosen, why? How are they related directly to the proposed method? It would be useful if authors mention it in the section 6 (after Line 464). If would be useful if the proposed method is compared with minimum five other methods for effectiveness. (Please see: https://www.mdpi.com/1424-8220/23/3/1310, and, https://www.mdpi.com/2076-3417/9/2/238)

4. Kindly insert a framework of the proposed model in graphical form, that will be useful. Please see: https://www.researchgate.net/publication/281445891_Clustering_the_Mixed_Numerical_and_Categorical_Dataset_using_Similarity_Weight_and_Filter_Method.

5. References are just adequate; it would be useful to include more methodical studies for better comparison and analysis.

Author Response

Point 1. The proposed model for real time data is based on a synthetic data set. It would be useful if application of the method is done on a real-world data set by creating a separate section (Assessing the performance, (please see: https://www.nature.com/articles/s41598-021-83340-8) and mentioning the results and model accuracy after that.

Response 1. The authors have incorporated the experimental study with a real world dataset, including performance, accuracy assessment as suggested by respected reviewer. Also, the paper given in link above is considered in the experimental study.

Point 2. The study is conducted on a synthetic dataset (Line 459). Perhaps, a paragraph on limitations of the study must be included in the conclusion section, or create another section of limitations of the study, whichever is suitable.

Response 2. The limitation of the study is included in the last section (7.2) along with lines for future works as suggested by respected reviewer.

Point 3. For comparative studies only two methods are chosen, why? How are they related directly to the proposed method? It would be useful if authors mention it in the section 6 (after Line 464). If would be useful if the proposed method is compared with minimum five other methods for effectiveness. (Please see: https://www.mdpi.com/1424-8220/23/3/1310, and, https://www.mdpi.com/2076-3417/9/2/238).

Response 3. For comparative studies five different approaches of similar nature are taken as suggested by the respected reviewer. These models are unsupervised follow either partitioning approach or hierarchical approach. The paper given in the above link has gone through and detailed analysis of 5 methods and comparative performance with proposed method is given in section-6.

Point 4. Kindly insert a framework of the proposed model in graphical form, that will be useful. Please see: https://www.researchgate.net/publication/281445891_Clustering_the_Mixed_Numerical_and_Categorical_Dataset_using_Similarity_Weight_and_Filter_Method.

Response 4. A flowchart describing the framework of the method is incorporated as suggested by the respected reviewer.

Point 5. References are just adequate; it would be useful to include more methodical studies for better comparison and analysis.

Response 5. Some latest references are included incorporating more methodological study in the introduction.

Finally, the authors gratefully acknowledge the helpful comments and suggestions of the reviewers, which have improved the papers and ready to do more if requires.

Reviewer 2 Report

Title of the paper: "A Mixed Clustering Approach for Real Time Anomaly Detection"

As a researcher working in the same field, I am impressed by the technique introduced in the paper, because it sheds new light on the earlier results of several authors and obviously can be successfully used in practice. From this point of view, the subject of the paper fits well with the scope of the journal (Applied Sciences).

The paper is ended with numerical simulations that corroborate the theoretical results.

This manuscript contains new ideas and good results that help other researchers.

The decision is too minor revision it for publication in the "Applied Sciences".

Therefore, I recommend publishing this work after taking these points into account.

1- Introduction needs to explain the main contributions of the work more clearly.

2- The novelty of this paper is not clear. The difference between present work and previous Works should be highlighted.

3- In the references in the Introduction section, the authors cite some works. However, they have not indicated the advantage or disadvantage and their relations to this paper. It’s a little confusing.

4- Comparison with recent studies and methods would be appreciated.

5- The authors can add the following reference to enrich the introductory section:

*A numerical method for solving the nonlinear equations of Emden-Fowler models, Journal of Ocean Engineering and Science, 2022. doi.org/10.1016/j.joes.2022.04.019.

*Stability, existence, and uniqueness for solving fractional glioblastoma multiforme using a Caputo–Fabrizio derivative, Mathematical Methods in the Applied Sciences, 2023. doi.org/10.1002/mma.9038

6- Future recommendations should be added to assist other researchers to extend the presented research analysis.

Sincerely Yours

Prof. Dr. Amr Mahdy

Author Response

Point 1. Introduction needs to explain the main contributions of the work more clearly.

Response 1. Introduction section is rewritten explaining the main contribution of the study as suggested by the respected reviewer.

Point 2. The novelty of this paper is not clear. The difference between present work and previous Works should be highlighted.

Response 2. The novelty of the paper is clearly explained and the differences between the proposed work and previous are highlighted in the introduction as suggested by the respected reviewer.

Point 3. In the references in the Introduction section, the authors cite some works. However, they have not indicated the advantage or disadvantage and their relations to this paper. It’s a little confusing.

Response 3. In introduction section is revised and most of the cited papers advantages and disadvantages are given in relation to the proposed work as suggested by the respected reviewer.

Point 4. Comparison with recent studies and methods would be appreciated.

Response 4. Comparison with recent studies and proposed methods are given incorporating more experimental analysis as suggested by the respected reviewer.

Point 5. The authors can add the following reference to enrich the introductory section:

*A numerical method for solving the nonlinear equations of Emden-Fowler models, Journal of Ocean Engineering and Science, 2022. doi.org/10.1016/j.joes.2022.04.019.

*Stability, existence, and uniqueness for solving fractional glioblastoma multiforme using a Caputo–Fabrizio derivative, Mathematical Methods in the Applied Sciences, 2023. doi.org/10.1002/mma.9038

Response 5. Above two references suggested by the respected reviewer are added.

Point 6. Future recommendations should be added to assist other researchers to extend the presented research analysis.

Response 6. According to the suggestion of the respected reviewer, future recommendations are added and given along with limitations in section-7.2.

Finally, the authors gratefully acknowledge the helpful comments and suggestions of the reviewers, which have improved the papers and ready to do more if requires.

Reviewer 3 Report

Abstract, title and references

(1) The abstract should follow the style of structured abstracts: 1) Background: Place the question addressed in a broad context and highlight the purpose of the study; 2) Methods; 3) Results; 4) Conclusion. Abstract came out well but it doesn’t have the statistical information such as how these experiment results solved the issue and what the percentage error in comparison to different solutions is.

(2) After going through references from 2021-2022 there is not a single publication.

(3) Authors should not use terms like: we, our

For example, your Abstract includes:

In this article, a mixture of partitioning and agglomerative hi10 erarchical approach to detect anomalies from such datasets has been introduced.

It should be written as:

In this article, a mixture of partitioning and agglomerative hi10 erarchical approach to detect anomalies from such datasets has been introduced.

Any parts that contain this style should be revised.

Introduction, Related Work

(1) Introduction section came out well provided that there is a clear substantiation on the gap or limitations in the existing methods.

(2) Total of 38 article citations are included. Out of which in introduction there are 36. Once again: After going through references from 2021-2022 there is not a single publication !!! A lot of citations to old literature.

(3) Authors should not use terms like: our

Problem Definitions, Proposed Algorithm, Time Complexity

- math formulas should be better organized (this applies to the entirety of this section)

- a block diagram should be drawn for the proposed algorithm

- the algorithm presented in the form of a list of steps should be better organized

- the calculations in the Time Complexity subchapter should be presented in a clearer way

Experimental Settings and discussions

(1) Maybe you need to rethink the title of the chapter? -> Experimental Analysis and Results ?

(2) Authors should not use terms like: our (change as previously described)

Conclusions

The discussion section lags in explanation with respect to the work carried out.

- There is no proper representation of the work with validation and justification, hardly articles are being used in discussion section.

- Conclusion looks to be generic need to compile the outcomes and state based on the tests conducted and convey how best this can fit in the current context.

- In conclusion section, values have to be displayed with explanation. It's better to mention the salient features of the entire work in terms of bullet points with current context.

Overall

The topic is relevant to current context but needs to incorporate all the modifications suggested above.

Author Response

Point 1. The abstract should follow the style of structured abstracts: 1) Background: Place the question addressed in a broad context and highlight the purpose of the study; 2) Methods; 3) Results; 4) Conclusion. Abstract came out well but it doesn’t have the statistical information such as how these experiment results solved the issue and what the percentage error in comparison to different solutions is.

Response 1. The abstract is revised incorporating all the aspect as well as statistical results as suggested by the respected reviewer.

Point 2. After going through references from 2021-2022 there is not a single publication.

Response 2. Latest references of 2022-23 are added as suggested by the respected reviewer.

Point 3. Authors should not use terms like: we, our

For example, your Abstract includes:

In this article, a mixture of partitioning and agglomerative hi10 erarchical approach to detect anomalies from such datasets has been introduced.

It should be written as:

In this article, a mixture of partitioning and agglomerative hi10 erarchical approach to detect anomalies from such datasets has been introduced.

Any parts that contain this style should be revised

Response 3. The paper is revised thoroughly and term “we”, “our” are removed by changing the sentences into passive form as suggested by the respected reviewer.

Point 4 Introduction, Related Work

4.1 Introduction section came out well provided that there is a clear substantiation on the gap or limitations in the existing methods.

Response 4.1. Introduction section is revised incorporating the limitations of the existing methods as suggested by the respected reviewer.

4.2 Total of 38 article citations are included. Out of which in introduction there are 36. Once again: After going through references from 2021-2022 there is not a single publication!!! A lot of citations to old literature.

Response 4.2. All total 45 article citations are included all are referred in Introduction and some articles of 2022-2023 are also cited.

4.3. Authors should not use terms like: our

Response 4.3. The term “we/our” are replaced using passive form of sentences.

Point 5. Problem Definitions, Proposed Algorithm, Time Complexity

- math formulas should be better organized (this applies to the entirety of this section)

- a block diagram should be drawn for the proposed algorithm

- the algorithm presented in the form of a list of steps should be better organized

- the calculations in the Time Complexity subchapter should be presented in a clearer way

Response 5.

Math formulae are written using MS word equation editor, authors have tried to organize or reorganize most of them as much as possible.

A block diagram is drawn as flowchart incorporating all the steps of the algorithm.

The time complexity calculation is presented stepwise manner. Each step’s cost given with justification to make it clear. The authors have tried to incorporate the suggestions of the respected reviewer.

Point 6. Experimental Settings and discussions

1. Maybe you need to rethink the title of the chapter? -> Experimental Analysis and Results?

Response 6.1. Experimental setting and discussion is changed to “Experimental Analysis and Results” as suggested by the respected reviewer.

2. Authors should not use terms like: our (change as previously described)

Response 6.2. The authors have removed terms like ‘our’ as suggested by respected reviewer.

Conclusions

The discussion section lags in explanation with respect to the work carried out.

- There is no proper representation of the work with validation and justification, hardly articles are being used in discussion section.

- Conclusion looks to be generic need to compile the outcomes and state based on the tests conducted and convey how best this can fit in the current context.

- In conclusion section, values have to be displayed with explanation. It's better to mention the salient features of the entire work in terms of bullet points with current context.

Response 7.

7.1. The discussion is revised incorporating most of the suggestions of the respected reviewer by adding proper explanations, validations, justifications, with respect to the work carried out. Comparative analysis with five other similar works are done.

2. The conclusion is revised making it concise incorporating the main features of the method along with outcomes using bullet points.

Finally, the authors gratefully acknowledge the helpful comments and suggestions of the reviewers, which have improved the work and ready to do more if requires.

Round 2

Reviewer 1 Report

I have no further comments. Thank you.