Relations between average shortest path length and another centralities in graphs

Mikhail Tuzhilin

2024-12-07

Abstract:Relations between average shortest path length and average clustering coefficient, radiality, closeness and stress centralities were obtained for simple graphs.

Social and Information Networks,Combinatorics

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to explore the relationship between the average shortest path length in a graph and other centrality indicators (such as radiality, closeness, stress centrality) as well as the average clustering coefficient. Specifically, the author focuses on the mathematical relationships among these quantities in simple connected graphs and attempts to establish connections between them under specific conditions.

### Main problems of the paper:
1. **Explore the relationship between different centrality indicators and the average shortest path length**: Researchers hope to understand how the average shortest path length is correlated with other centrality indicators such as radiality, closeness, and stress centrality.
2. **Explore the relationship between the average shortest path length and the average clustering coefficient**: Pay special attention to the relationship between the average shortest path length and the average clustering coefficient for geodetic graphs or graphs with odd - length cycles under certain additional conditions.
3. **Verify the characteristics of small - world networks**: By analyzing these relationships, further verify whether the network has small - world characteristics, that is, a relatively small average shortest path length and a relatively large average clustering coefficient.

### Key contributions of the paper:
- **Establish the relationship between local shortest path length and local clustering coefficient**: Prove that the relationship between the local shortest path length \(L(N(i))\) and the local clustering coefficient \(c_i\) is \(L(N(i)) = 2 - c_i\).
- **Explore the relationship between the shortest path length in the induced subgraph and the global shortest path length**: Propose several theorems and lemmas to describe the change rules of the shortest path length after deleting a vertex or adding a vertex.
- **Establish the relationship between the average closeness centrality and the average shortest path length**: Prove that the average shortest path length \(L(G)\) is not less than the sum of the average closeness centrality \(Clo(v)\) of nodes.
- **Establish the relationship between the average radiality and the average shortest path length**: Give the relationship formula between the average shortest path length \(L(G)\) and the average radiality \(Rad(v)\) as \(L(G)=\text{diam}(G)+1-\frac{1}{n}\sum_{v\in V(G)}Rad(v)\).
- **Explore the relationship between stress centrality and the average shortest path length in geodetic graphs**: For geodetic graphs, establish the relationship between stress centrality \(Str(i)\) and the average shortest path length \(L(G)\).

### Formula summary:
1. **Relationship between local shortest path length and local clustering coefficient**:
\[
L(N(i)) = 2 - c_i
\]
2. **Relationship between average shortest path length and average closeness centrality**:
\[
L(G)\geq n\sum_{v\in V(G)}Clo(v)
\]
3. **Relationship between average shortest path length and average radiality**:
\[
L(G)=\text{diam}(G)+1-\frac{1}{n}\sum_{v\in V(G)}Rad(v)
\]
4. **Relationship between stress centrality and average shortest path length in geodetic graphs**:
\[
L(G)=1+\frac{1}{n(n - 1)}\sum_{i\in V(G)}Str(i)
\]

Through the research of these relationships, the paper provides a theoretical basis for understanding complex network structures and their characteristics.