Bibliometric Mining of Research Trends in Machine Learning

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The figure on Figure 4 does not have a figure number, and could be organized in a more visual way. 

In addition, this paper is very similar (although different) to the following paper by the author, and in this sense, I believe that the contribution to the literature has been made by the paper mentioned in the references below. 

 

Lundberg, L. Bibliometric mining of research directions and trends for big data. J Big Data 10, 112 (2023). https://doi.org/10.1186/s40537-023-00793-6

 

Author Response

The response is in the attached word file "Response to Reviewer 1"

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

The research presents a method for bibliometric mining of research trends in machine learning, analyzing productivity, growth rate, and citations for different research directions. The methodology involves creating a hierarchical taxonomy of research directions in machine learning, with experts using a Python program to extract common keywords from the Scopus database. The study covers a large corpus of 398,782 documents and identifies the largest and fastest-growing research directions, such as Algorithms and Applications, including sub-directions like Convolutional Neural Networks and Internet of Things. The analysis also explores the geographic distribution of research in machine learning. The approach is semi-automatic, combining expert-defined taxonomies with data mining, and aims to adapt to emerging topics and new trends in dynamic research areas. I would suggest revisions as listed below:

1.      What are the most influential authors and journals in the machine learning research area, and how have they evolved over time?

2.      What are the emerging topics and trends in machine learning research, and how can they be identified and analyzed using bibliometric methods?

3.      How can bibliometric mining of research trends in machine learning be used to inform policy and decision-making in industry and academia?

4.      Many recent research works use multi-modal machine learning methods such contrastive learning. The authors should consider these cutting-edge approaches. Some references to build the related work

https://arxiv.org/abs/2305.19894

https://aclanthology.org/2022.acl-demo.3/

 

 

Comments on the Quality of English Language

english is fine

Author Response

The response is in the attached Word file (Response to Reviewer 2).

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

In this paper, Lars Lundberg et al. used the bibliometric mining method to examine research trends in machine learning. The paper is well-written and demonstrates the authors' knowledge background. It also brings benefits to the scientific world and contributes to the machine learning research field. This paper will generate a lot of research interest. As a result, I recommend this paper be published after several minor modifications. Below are my suggestions:

Line 27: "The basic idea in Machine Learning (ML) is that the behavior of a computer..." Currently, the authors use "machine learning" and "ML" multiple times in the paper. If the authors prefer "ML," they should use it consistently throughout the paper. If the authors prefer "machine learning," there is no need to introduce "ML."

Line 92: Change "Related Works" to "Literature Reviews."

Lines 203, 205, 244: The authors used "Fig.1" and "Fig.2." However, in Lines 256 and 259, the authors used the caption as "Figure 1" and "Figure 2." Consistency is recommended for the entire paper. Please check before publication.

Line 352: Please check whether it is "Linear" or "Linear regression."

Line 359: For Figure 6, check the data for "Learning paradigms." The numbers for 2019 and 2020 appear to be the same.

Line 399: For Figure 14, clarify the meaning of the horizontal axis. The authors need to provide the legend and the unit.

Lines 437 and 461: For Figure 17 and Figure 21, I suggest including the full spellings of "NCS" in the captions to assist readers and researchers

Author Response

The response is in the attached Word File (Response to Reviewer 3)

Author Response File: Author Response.pdf

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors, 

the manuscript is interesting and very well written. The literature overview is very detailed and informative. I have some minor comments and suggestions that can slightly improve the manuscript's quality. 

1. If you have introduced the abbreviations in the manuscript at some point please use it from that point on. For example, ML (machine learning), convolutional neural networks (CNN), and Artificial Intelligence (AI) have been introduced at the beginning of the manuscript so you don't use full names in the Conclusions section. 

2. If the entire manuscript is written in Times New Roman font style please use it in Figures/graphs. 

Author Response

The response is in the attached Word file (Response to Reviewer 4)

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

I appreciate the authors' earnest efforts in addressing the reviewers' comments, which I found satisfactory. Nevertheless, I would like to propose a minor enhancement. It would be beneficial to organize the data collected since 2013 as a GitHub file and include it in the references section rather than presenting it as an appendix. This adjustment is suggested for the convenience of readers.

Author Response

We have followed the reviewer's recommendation and put all data on https://github.com/Lars-Lundberg-bth/bibliometric-ml. The manuscript has been updated accordingly, i.e., Appendix C has been removed and  all references to Appendix C in the text has been replaced with references to GitHub.