Detecting the Structural Hole for Social Communities Based on Conductance–Degree
Round 1
Reviewer 1 Report
Thank you for your contribution. I thought this was a well written interesting paper.
General Comments:
3rd Paragraph on Page 1: This seems to be the crux of the argument for this paper. I think it needs to be explained a little more clearly. I think you are trying to say that existing models identify important nodes in a network that are not necessarily structural hole spanners but just a central node. This can tie in better for the justification of the paper.
Explanation of the formulas for section 2.3 need to be cleaned up (see below).
“All of the teams were divided into 12 groups.” Page 7 – what are the 12 groups? Why?
Figure 5 – I think to compare these it might be useful to force the same layout, or the same position for each node. Then you can see how the different algorithms place nodes in the same\different community and make it easier to see if the same structural hole spanners are identified (or not, and why they might be different). Why not create a similar figure for the football network?
You mention a synthetic dataset on Page 2, but I didn’t see anything about this. It is only mentioned again in section 5. What happened to the results here?
Editorial Comments:
I think equations 1 and 2 have some formatting problems. Is it S_bar supposed to be a subscript? D_sum
S_bar is not defined. Is that supposed be T?
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
In the paper authors presented: (i) the Conductance-Degree structural holes detection algorithm, which computes the conductance and degree score of a vertex to identify the structural hole spanners in social networks; and (ii) improved Label Propagation Algorithm based on conductance to filter the jamming nodes, which have a high conductance and degree score but are not structural holes. The experimental results show that the algorithm can detect the structural holes and communities accurately and efficiently.
But the article has several disadvantages:
1. Sections 3 and 4 have the same name - "Conductance-Degree Structural Holes Detection Algorithm"
2. It is too early at the beginning of section 3 to say "In this section, we propose an efficient and accurate algorithm to detect structural hole spanners" (page 4). As an introductory presentation of the algorithm, it is necessary to briefly formulate (i) how it will be qualitatively different from all previous ones, (ii) what shortcomings of previous developments it will eliminate, (iii) which properties it will expand / change, etc.
3. It is desirable to determine what definition of the concept of a Structural holes the authors use in their research, and what specific properties they should possess, unlike other Social network nodes. Since the literature review (section 2.1.) gives several concepts proposed by different authors, but there is no single concept.
4. It is desirable to add a Discussion section for comparisons the differences between the results obtained using the proposed algorithm and those used previously.
5. There is no information about the Limitations of the proposed algorithm
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
In the paper authors presented: (i) the Conductance-Degree structural holes detection algorithm, which computes the conductance and degree score of a vertex to identify the structural hole spanners in social networks; and (ii) improved Label Propagation Algorithm based on conductance to filter the jamming nodes, which have a high conductance and degree score but are not structural holes. The experimental results show that the algorithm can detect the structural holes and communities accurately and efficiently.
Paper could be recommended for publication due to authors revision.
