A Novel Process Recommendation Method That Integrates Disjoint Paths and Sequential Patterns

Round 1

Reviewer 1 Report

- Add a flowchart to explain the research methodology/design

- Algorithms 1,2 and 3 are not formatted according to a uniform pattern, some have numbering for each step while others don't. Make sure all the algorithms are uniformly written (writing convention, numbering/bulleting, uppercase/lowercase all should be aligned). Moreover, explain the meaning of each symbol/denotation used wherever applicable.

- Why did you chose this dataset? What is the name? Where did you get it from? Cite the source and provide strong reasoning of choosing this dataset as this is very important to decide if your results are worthy of being accepted.

- Explain real life/daily life implications/applications that will be benefitted by your research.

- Elaborate your future work further. Try to answer the following questions to further enhance it:

   1. What are limitations of your work? What are the areas/weaker aspects that still need further explored for betterment of research community and industry? How can readers tackle those issues in future (basic guideline)?

- Most of the references are outdated, giving an impression that this study is not the need of current time. Update/add references of recent years.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The work is excellent and very useful for both research and business management.
However, I think it would be improved with:
1. a revision in the wording of 1.Introduction and 2. Related Work. For there are abundant redundancies in the elaboration of the paragraphs and the use of adverbs.
2.    Eliminating 3.Related Concepts, since it deals with basic terms and definitions, which are only useful in teaching papers, not in academic research.

Author Response

Please see the attachment

Author Response File: Author Response.docx

Reviewer 3 Report

In general, the topic is interesting and up-to-date.

The content of the manuscript is presented in a logically consistent order. However, the article lacks in its structure a separate "Discussion and limitations" section. 

The "Related work" section is too modest. The literature review lacks the aspect of optimizing the execution of business processes and the possibility of their dynamic reconfiguration. Please refer to the following up-to-date articles: "Optimization of Business Process Execution in Services Architecture: A Systematic Literature Review" (https://0-doi-org.brum.beds.ac.uk/10.1109/ACCESS.2021.3102668), "Edge Intelligence Service Orchestration with Process Mining" (https://0-doi-org.brum.beds.ac.uk/10.3390/app122010436), "Investigation of Microservice-Based Workflow Management Solutions for Industrial Automation" (https://0-doi-org.brum.beds.ac.uk/10.3390/app13031835). Topics should also be commented on in the "Discussion and limitations" section in relation to the proposed process recommendation method.

Besides, in the "Related work" section in line 131 the authors state "However, the results are not satisfactory". Please elaborate more on those results.

The "Proposed Method" section needs to be refined. In particular, both methods (subsection 4.1) "Edge-disjoint path extraction method" and (subsection 4.2) "CPSP mining method" require greater precision in the description. Method diagrams should be improved. For example, in Figure 4, there are no actions after the two decision points. However, in Figures 5 and 6 there are no decision points and returns to previous actions are used. This requires decision points or lines of synchronization of parallel actions. The article concerns processes, so it would be appropriate to use, for example, a UML activity diagram to present the flow of activities in the algorithms.

In the "Discussion and limitations" section the authors should discuss the pros and cons of the methods. That is the place for a wider analysis of the results and drawing general findings. Practical applications of the developed method should also be presented and commented on.

It is worth elaborating more on the planned further works in the "Conclusions" section.

The writing style of the article needs improvement. For example, the sentences "reliable recommendations for process recommendations" (line 68) and "process modeling process" (line 75) should be rewritten. Please check the whole manuscript.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

In general, the topic is up-to-date.

The content of the manuscript is presented in a logically consistent order.

I confirm that the authors have addressed many of my concerns. However, there are a few changes that the authors must incorporate prior to publication.

The authors added the "Discussion and limitations" section and significantly improved the "References" section according to all comments.

However, the aspect of optimizing the execution of business processes and the possibility of their dynamic reconfiguration is not fully covered. The authors used two suggested papers but missed an important one that considers the reliability of components/services - https://0-doi-org.brum.beds.ac.uk/10.1109/ACCESS.2021.3102668. The comments on the topic should be augmented in the "Discussion and limitations" section in relation to the proposed process recommendation method.

I appreciate that the authors have used the Unified Modeling Language (UML) Activity diagrams to present algorithms. But, they should also use the UML Activity diagram in figure 2. In its current form, the presented flow is not precise enough.

Author Response

Please see the attachment.

Author Response File: Author Response.docx