Building Political Hashtag Communities: A Multiplex Network Analysis of U.S. Senators on Twitter during the 2022 Midterm Elections

Round 1

Reviewer 1 Report

In my opinion, it is a well-written article. The paper has novel results that I find quite interesting.

My comments are given below:

1) How you evaluate the quality of ChatGPT clustering?

2) The time span you investigate includes the election. In this regard, one may hypothesize that behavior patterns before and after the election can differ. Please address this point.

3) Table 1. The last column (# of clusters). What are these clusters about? Please give more detailed explanation. In Subsection 4.2, the authors talk about hashtag clustering. If the corresponding column stands for hashtag clusters than I do not fully understand why the number of clusters is of the same order than the number of hashtags. Do we need such a clustering?

4) Please introduce Jeffrey degree and Pearson degree (if I am correct, the capital letters are needed - see lines 189, 190).

5) Figure 3. I suppose that for effective comparing of distributions, one need to extend the figure by adding relative quantities. That is, what is the fraction of the edge Economy among the Democrat party? Further, I suppose that proportion confidence intervals should be included.

6) Figure 4. Please add a color bar and the correlation values.

Author Response

All responses are written in the attached document.

Author Response File: Author Response.pdf

Reviewer 2 Report

The following are my comments for improvement of this paper:

1. The authors state – “The supervised multiplex network has seven layers and the unsupervised one has six.” Why is it necessary to have seven layers in the supervised multiplex network and six in the unsupervised one?

2. The authors state – “Data used to construct this network was gathered from a period spanning one month before and one month after the election.” – this statement is unclear and is missing essential details. Please re-write this section to clearly discuss the step-by-step methodology that was followed for data collection. Specifically, the authors should state whether they used the Standard Search API or the Advanced Search API of Twitter. As the Twitter API has rate limits, this section should also present the step(s) that were taken to address the rate limits.

3. The authors state – “Do politicians strategically use hashtags to build their own communities? To address this question…..” Why is hashtag analysis relevant and important in the context of data analysis of social media conversations? Please elaborate the same by briefly reviewing some recent works in hashtag analysis.

4. Section 4.2 is poorly written. It seems ChatGPT was used like a black box where the data was provided and the results were obtained. Why is it necessary to use ChatGPT for the kind of analysis that is reported?

5. The presentation of the methodology is not clear. What are the characteristics of the Multiplex Network Analysis? 

6. A comparison with prior works is missing: Please include a comparative study (qualitative and quantitative) with prior works in this field to highlight the novelty of this work. 

Author Response

Responses are in the attached document. 

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The authors addressed all of my concerns. I have no further comments.

Good luck!

Reviewer 2 Report

The authors have revised their paper as per my comments and feedback. I do not have any additional comments at this point. I recommend the publication of this paper in its current form.