Rough IPFCM Clustering Algorithm and Its Application on Smart Phones with Euclidean Distance

Round 1

Reviewer 1 Report

Thank you for inviting me as a reviewer for a manuscript titled Rough IPFCM Clustering Algorithm and Its Application on Smart Phone with Euclidean Distance. The paper is impressive for the efforts made by you to demonstrate the valence of your model. The model is well explained and the methodology is clear. But, the paper would be more exciting if you implement the below improvements:

. Need to better highlight the novelty of the study in the introduction. Why is the topic important (or why do you study it)? What are the research questions? What are your contributions? Why is it to propose this particular method (IPFCM))? What is the power of the proposed algorithms that you are exploring in this research?

. Better define the motivations.

. Literature review. The rough set, fuzzy sets and C-means are widely used for engineering optimisations, solving various management problems. Authors should provide some recent relevant references (from 2019-2022) like: Sahu , R., Dash , S. R., & Das, S. (2021). Career selection of students using hybridized distance measure based on picture fuzzy set and rough set theory. Decision Making: Applications in Management and Engineering, 4(1), 104-126. https://doi.org/10.31181/dmame2104104s; Sharma, H. K., Kumari, K., & Kar, S. (2020). A rough set approach for forecasting models. Decision Making: Applications in Management and Engineering, 3(1), 1-21. https://doi.org/10.31181/dmame2003001s.

. Generally, validation and comparisons of the results is well prepared.

. The conclusion section -  The authors will have to demonstrate the impact and insights of the research. Add limitations of the model.  
Scientific soundness :
. The subject addressed in this paper is relevant.  
Interest to the readers :
. In my opinion, method of this paper seem to be interesting for the readership of the journal.

Author Response

See attached file.

Author Response File: Author Response.pdf

Reviewer 2 Report

This paper proposed an interval clustering framework for symbolic data analysis on smartphones. It combines the rough set and interval possibilistic fuzzy C-means using the Euclidean distance. The experiment was conducted on four cases with different situations such as containing outliers and noises. The proposed algorithm was compared with the other two algorithms in terms of RMSE.

In general, the topic conducted in the paper is interesting and the results seem to be reasonable. However, it has many issues that the authors need a carefully major revision to improve the quality of the paper. My detailed comments are as follows.

First, the paper needs proofreading, preferable by native speakers to improve the writing style. The current text is very confusing, long, and redundant.

The authors need to pay much attention to rewriting the Abstract and Introduction. There's a lot of redundancy I could find. For instance, the authors stated that they proposed the IPFCM, then IFCM, then later RIPFCM, but they explained in a way that they dismissed what they have proposed. Why did you propose IFCM even though it has limitations?

Again, in the Introduction, I can not see clearly the motivation and contribution of the paper. Instead, the authors introduced several previous works and then repeat some points that were already stated in the Abstract. Thus, it is better to highlight the contributions. Move some paragraphs to a new Section called Related work.

In the Related work, introduce several works in this field, emphasis on their limitations.

In section 2, insert a table of notations used in the paper.

Also in this section, I suggest authors use a demo dataset to make examples for each definition in this section. It is better for readers to follow instead of plain text and formulations.

The pseudo-code in Algorithm 1 should be revised in a form of an algorithm, the authors can follow the conventional way to make an Algorithm from [https://en.wikibooks.org/wiki/LaTeX/Algorithms#/media/File:Latex-algorithm2e-if-else.png]

Section 3 is much better than sections 1 and 2. However, I suggest authors use other metrics such as the Silhouette coefficient to verify and visualize the result. It is feasible since computing the Silhouette you only need a distance matrix of pairwise instances which is easily obtained by using Euclidean distance in your work.

Give the reasons for the choice of the number of clusters in each case.

In the discussion and conclusion, the authors should discuss possible clustering methods that not only can perform clustering tasks well but also can enhance the interpretability of the results. A good option is Hierarchical clustering. The authors refer to several good examples[https://doi.org/10.1007/978-981-15-1209-4_1]and [https://doi.org/10.3390/app112311122] in the discussion.

 

 

 

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Reviewer 3 Report

After careful review, the manuscript has a reasonable effort and technical information.  However, there are some points that must be considered in the revision to be worth being accepted. Therefore, I strongly recommend the authors to follow the comments below:

1- As you have some repetitive abbreviations, therefore please provide an abbreviation table that on MDPI format must be at the end of the paper, 

2- Abstract: The sentence is long and a bit confusing. You can concise it and point the important parts and highlights the achievements and what makes this paper special.

3- The work presents a poor literature review on the methods based on soft computing (ML, Fuzzy and MCDM) for this purposes and some smartphone application for different purposes. Here you can find some of the new works you can use on ur study:

-Evaluation of Machine Learning and Web-Based Process for Damage Score Estimation of Existing Buildings

-Precision irrigation system (PIS) using sensor network technology integrated with IOS/Android application

-Application of iOS/Android based assessment and monitoring system for building inventory under seismic impact

-ML-EHSAPP: A prototype for machine learning-based earthquake hazard safety assessment of structures by using a smartphone app

4- Please improve the quality and font size of figures to be more visible

5- Please provide a figure that shows the architecture of your model and network.  Something like a graphical abstract.

6- Please make your tables and figures follow a same path and font size and colors and adjust them in a proper way.

7- There are many typos that needs to be corrected, font size, table formats, equations, legends in the figures. Please make the format similar for all. Also, be careful about spaces after (.) amd (,) which in some cases are doubled or missed. Kindly value your works and time you spend to write and submit, this is not really professional way of presenting.

8- You did not highlight the problem statement, objectives and novelty of your proposed method; That is why increasing the background of literature review based on the recommended works can help in this manner.

9- There is a need in strong proofreading of the work.

10- Please provide more information about the figures and tables and write more in the body of the text about them and the information they provide.

11- Your conclusion is very short and it is not scientificly presented.

12- It would be great if you do a comparison between existing methods and your methods and what is the novelty of your work?

13- Please give more information about data repository and case study. Or provide a numerical example to make it easier for readers to understand.

At the end as I have mentioned, there are not much significant novelty on this work but the efforts were good and it would be good if you revise it  totally according to the points provided and other reviewers.

Author Response

Please see the attached file.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The authors have revised the paper following most of my suggestions.

In this version, the authors should further improve its quality before I vote for an acceptance.

- Theoretically discuss the complexity of the proposed algorithms.

- Put the references for two works that I recommend to you in the previous review.

 

Reviewer 3 Report

Dear Authors

Many thanks for significant changes and responses.