Next Article in Journal
A Combined System Metrics Approach to Cloud Service Reliability Using Artificial Intelligence
Next Article in Special Issue
Operations with Nested Named Sets as a Tool for Artificial Intelligence
Previous Article in Journal
A Framework for Content-Based Search in Large Music Collections
Previous Article in Special Issue
Infusing Autopoietic and Cognitive Behaviors into Digital Automata to Improve Their Sentience, Resilience, and Intelligence
 
 
Article
Peer-Review Record

Combination of Reduction Detection Using TOPSIS for Gene Expression Data Analysis

Big Data Cogn. Comput. 2022, 6(1), 24; https://doi.org/10.3390/bdcc6010024
by Jogeswar Tripathy 1, Rasmita Dash 1,*, Binod Kumar Pattanayak 1, Sambit Kumar Mishra 2,*, Tapas Kumar Mishra 2 and Deepak Puthal 3,*
Reviewer 1: Anonymous
Reviewer 2:
Big Data Cogn. Comput. 2022, 6(1), 24; https://doi.org/10.3390/bdcc6010024
Submission received: 27 December 2021 / Revised: 13 February 2022 / Accepted: 16 February 2022 / Published: 23 February 2022
(This article belongs to the Special Issue Data, Structure, and Information in Artificial Intelligence)

Round 1

Reviewer 1 Report

The authors presented an interesting comparison of various techniques of performing feature selection in machine learning algorithms for cancer gene expression data. The topic is important and should be of interest. 

I have some suggestions on the description of the methods and presentations, which I hope that the authors should improve. 

1) the acronyms should be defined before being used. Some acronyms are defined in the abstract but not in the main text, which should be added. The methods should be cited when it is first introduced. For example, line 74, "Four feature ranking algorithms are taken into account in this model, such as CBFS, CST, InG, and RFS for getting a better feature subset from datasets. Then the classification techniques such as k-NN, RF, LR, and DT are used to classify the microarray databases.". None of the acronyms are defined before they are used, and no reference is cited. This is hard to interpret even for experts. 

2. There are so many acronyms and it would be nice to make a table of them. I find it very hard to follow. 

3. The authors suggest that the method may work well for some datasets but not others. Indeed, whether a method works well depends on the truth of the data, which is not known. Can the authors comment on how generalizable these methods are? 

4. The authors discussed their methods in the context of microarray gene expression. How generalizable is the method to other diseases and other types of data (e.g., RNA-seq, methylation data). Can the authors comment? 

Author Response

Dear Reviewer,

I am attaching the reviewer to response file.

Author Response File: Author Response.pdf

Reviewer 2 Report

paper is well presented. results are easy to follow and interpreted. Here are some comments:

  1. the proposed model in fig 1.2, should be more explained a bit in the proposed work. Can be confusing for the reader to make a distinction in which is the contributing part. 
  2. Future work is not clear to specific applications that can benefit from this research.

Author Response

Dear Reviewer,

I am attaching the reviewer to response file.

Author Response File: Author Response.pdf

Back to TopTop