Experimental Analyses of Temporal Activity-Sequencing Anomalies in Process Mining
Round 1
Reviewer 1 Report
I congratulate you for this work,
my remarks are the following:
Related works should follow a logical and chronological order trying to make leavings between a work and its successor.
Review the presentation of the "Related works" part in the form of points ...
Figure 1 "Activity-Sequencing Graph in the Process Mining Framework Dealing with Temporal Activity-Sequencing Anomalies" is not explained.
the number of references is low compared to the complexity of the research topic
Good luck
Author Response
1) Related works should follow a logical and chronological order trying to make leavings between a work and its successor. Review the presentation of the "Related works" part in the form of points
I revised the Related works so as to describe the works with chronologically ordered items.
2) Figure 1 "Activity-Sequencing Graph in the Process Mining Framework Dealing with Temporal Activity-Sequencing Anomalies" is not explained.
I fully revised the subsection of the framework by adding the paragraphs.
3) the number of references is low compared to the complexity of the research topic.
I revised the references by adding 16 references.
Reviewer 2 Report
The investigated topic is interesting. The author divides the manuscript into theoretical and experimental parts (these section are exhaustive), also showing the tests performed. However, this reviewer did not find an exhaustive explanation of why this method is better than those already present in the literature, a comparison that makes us understand the scientific contribution on the state of the art. This part needs to be added.
Some typos:
This is manuscript is a single author so in the text should not use “we” or “our” …
It is necessary to check the links to the bibliography, some are skipped such as in the third line of the "Related Works" section or page 6 “The algorithm ?? describes … ”
Figure 6 should be better arranged
It would be better if the bibliography contains at least 20 references
Proofreading by a native speaker is suggested
Author Response
1) The investigated topic is interesting. The author divides the manuscript into theoretical and experimental parts (these section are exhaustive), also showing the tests performed. However, this reviewer did not find an exhaustive explanation of why this method is better than those already present in the literature, a comparison that makes us understand the scientific contribution on the state of the art. This part needs to be added.
Rebuttal) Please let me explain. I explained the reason of the anomalies caused from the temporal activity-sequencing noises in the subsection of 3.1 and 3.2. Additionally, I would remind the reviewer that the point of this paper is not comparing with others' but detecting a new and specific type of noises.
2) Some typos:
This is manuscript is a single author so in the text should not use “we” or “our” …
I revised all the sentences starting with "we" or "our".
It is necessary to check the links to the bibliography, some are skipped such as in the third line of the "Related Works" section or page 6 “The algorithm ?? describes … ”
I fixed all the missing figure/algorithm references.
Figure 6 should be better arranged
I rearranged all the figures including Figure 6.
It would be better if the bibliography contains at least 20 references
I added additional 16 references.
Proofreading by a native speaker is suggested
I revised the overall sentences and paragraphs and fixed all the typos.
Reviewer 3 Report
Figure 4-7 are the key results from experiment, and are clearly presented. In addition to a general explanation, a more detailed explanation in Figure 6-7 will provide more evidences for the evaluation of the research. Conclusion can definitely be enhanced by the further issues discovered in this research.
Author Response
Figure 4-7 are the key results from experiment, and are clearly presented. In addition to a general explanation, a more detailed explanation in Figure 6-7 will provide more evidences for the evaluation of the research. Conclusion can definitely be enhanced by the further issues discovered in this research.
I revised several subsections related to the figures and enhanced the explanations. I appreciate the reviewer's positive and encouraging comments.
Reviewer 4 Report
In this work, the authors have explored the event-log anomalies and noises produced by the special type of anomalies inevitably formed in the event-log preprocessing phase of the automated process discovery. In order to further improve the quality of this manuscript, the following comments should be carefully considered:
1.The manuscript has some typos, ambiguous sentences and language issue that require full spell and syntax checking.
2.In the introduction, the authors have not surveyed the existing related references well and also some of the references are quite old. To improve the literature review, related published papers in recent years with stating advantages and disadvantage of them should be cited and the major contribution with regard to the existing works should be revised to highlight the innovation points of this article.
3.In order to make it easier for readers to grasp the key points, contributions to this manuscript should be further condensed. Significantly, some are work but not contributions.
4.Figure 2 is chaotic. Please check them carefully.
5.I would suggest that the authors must discuss some existing algorithms, which are widely utilized these days to solve several real world problems including the described issues: neuroadaptive learning algorithm for constrained nonlinear systems with disturbance rejection, asymptotic tracking with novel integral robust schemes for mismatched uncertain nonlinear systems, and so on.
6.In the fourth part, the authors should explain how to ensure the fairness of the compared methods.
7.The conclusions should be further condensed. And future research directions need to be included in the conclusion part.
8.The format of references needs to be further standardized.
Author Response
1.The manuscript has some typos, ambiguous sentences and language issue that require full spell and syntax checking.
I fixed all the typos and revised the overall sentences and paragraphs over the paper.
2.In the introduction, the authors have not surveyed the existing related references well and also some of the references are quite old. To improve the literature review, related published papers in recent years with stating advantages and disadvantage of them should be cited and the major contribution with regard to the existing works should be revised to highlight the innovation points of this article.
I fully revised the section of Introduction according to the reviewer's comments.
3.In order to make it easier for readers to grasp the key points, contributions to this manuscript should be further condensed. Significantly, some are work but not contributions.
I rearranged and revised the subsection of the framework to make clearer the scope as well as the contributions of this paper.
4.Figure 2 is chaotic. Please check them carefully.
I tried to depict the details of the noise detection and removal algorithm to make the reader much easier to grasp the proposed algorithm. But, I didn't explain the graphical representations of each step's output. The readers including the reviewer felt chaotic in the figure. With accepting the reviewer's comment I get rid of the figure.
5.I would suggest that the authors must discuss some existing algorithms, which are widely utilized these days to solve several real world problems including the described issues: neuroadaptive learning algorithm for constrained nonlinear systems with disturbance rejection, asymptotic tracking with novel integral robust schemes for mismatched uncertain nonlinear systems, and so on.
I would say that the proposed concept and its related algorithm ought to be the firstly raised issue in the literature. Therefore, it ought to be the only one and the other existing algorithms are not directly dealing with the activity-sequencing noises and anomalies. This is why I couldn't discuss and compare with others.
6.In the fourth part, the authors should explain how to ensure the fairness of the compared methods.
This is the same issue with the previous reviewer's comments.
7.The conclusions should be further condensed. And future research directions need to be included in the conclusion part.
I made the conclusion condensed by moving the sentences with a bullet to the proper part of the algorithm subsection.
8.The format of references needs to be further standardized.
I revised and fixed all the references improperly stated, and I added 16 additional references.
Reviewer 5 Report
Comments for author File: 
Comments.pdf
Author Response
From this point of view, the article is completely fine in my opinion. However, I would criticize the authors that the published algorithms are too small, have poor resolution, so they are often completely unreadable. They are therefore only an illustration for the reader, but do not fulfill the function of describing the algorithm used. I propose to enlarge the algorithms and add (at least some) as supplementary material at the end of the article.
I used the Latex-style of the MDPI applied sciences to format the algorithm proposed in the paper. Anyway, I revised the subsection related to the algorithm.
Furthermore, the following inaccuracies need to be removed: Reference number one is inaccurate. The names of the authors are wrongly given. The first author's name is completely missing.
I fixed all the references inaccurate and added 17 additional references including the article recommended by the reviewer.
Round 2
Reviewer 2 Report
All my suggestions have been applied.
