Degree and Clustering

Module 5: Complex Networks

Toolkit page

This page is self-study material for Session 23. It continues Paths in Directed Graphs.

This page computes degree and clustering with networkx (Newman 2018). You will need degree moments in the epidemic lab.

You reuse the two graphs from the previous pages.

import networkx as nx
import matplotlib.pyplot as plt

# Create an undirected graph
# (foods that go well together)
G_foods = nx.from_edgelist([
    ("bread", "cheese"),
    ("bread", "chocolate"),
    ("bread", "cured ham"),
    ("cheese", "cured ham"),
    ("chocolate", "cookies"),
    ("chocolate", "milk"),
    ("chocolate", "watermelon"),
    ("cookies", "milk"),
    ("cured ham", "watermelon"),
])

# Visualize the graph
pos = nx.circular_layout(G_foods)
nx.draw(G_foods, node_color="lightblue", with_labels=True, pos=pos)
plt.show()

# Create a directed graph
D_food_chain = nx.DiGraph()  # food chain (prey -> predator)
D_food_chain.add_edges_from([
    ("grass", "antelope"),
    ("grass", "zebra"),
    ("antelope", "lion"),
    ("antelope", "hyena"),
    ("zebra", "lion"),
    ("zebra", "hyena"),
    ("hyena", "lion"),
])

# Visualize the graph
pos = nx.spring_layout(D_food_chain)
nx.draw(D_food_chain, node_color="lightgreen", with_labels=True, pos=pos)
plt.show()

Degree

The degree \(k\) of a node is the number of edges connected to it. In an undirected graph, the degree of a node is simply the number of neighbors it has.

# Degree of a node
print("Degree of 'bread':", G_foods.degree("bread"))

Degree of 'bread': 3

The average degree of an undirected graph is given by:

\[\langle k \rangle = \frac{2E}{N}\]

where \(E\) is the number of edges and \(N\) is the number of nodes.

# Average degree of the graph
ls_degrees = [G_foods.degree(node) for node in G_foods.nodes()]
avg_degree = sum(ls_degrees) / len(ls_degrees)
print("Average degree:", avg_degree)

Average degree: 2.5714285714285716

Degrees in Directed Graphs

In a directed graph, you distinguish between in-degree (number of incoming edges) and out-degree (number of outgoing edges).

print("In-degree of 'antelope' (number of prey):", D_food_chain.in_degree("antelope"))
print("Out-degree of 'antelope' (number of predators):", D_food_chain.out_degree("antelope"))

In-degree of 'antelope' (number of prey): 1
Out-degree of 'antelope' (number of predators): 2

Clustering Coefficient

Many real networks have triangles: if A is connected to B and C, there is a higher-than-random chance that B is also connected to C (Watts and Strogatz 1998; Newman 2018). This tendency is captured by the clustering coefficient.

For an undirected graph, the local clustering coefficient of node \(i\) is:

\[ C_i = \frac{2T_i}{k_i(k_i - 1)} \]

where \(k_i\) is the degree of node \(i\), and \(T_i\) is the number of triangles that include node \(i\). If \(k_i < 2\), define \(C_i = 0\).

In the food graph, \(C_{\text{bread}} = 1/3\): “bread” has three neighbors, and only one neighbor-pair is connected.

The average clustering coefficient is:

\[ C = \frac{1}{N}\sum_{i=1}^N C_i. \]

In NetworkX:

• nx.clustering(G) returns \(C_i\) for every node.
• nx.average_clustering(G) returns the average \(C\).

You will use \(C\) later to compare random vs small-world vs scale-free synthetic networks.

# Local clustering coefficient for each node
clustering = nx.clustering(G_foods)
print("Local clustering:")
for node, c in clustering.items():
    print(f"  {node}: {c:.3f}")

print("Average clustering:", nx.average_clustering(G_foods))

# Another common global measure is transitivity
# (ratio of triangles to connected triples)
print("Transitivity:", nx.transitivity(G_foods))

Local clustering:
  bread: 0.333
  cheese: 1.000
  chocolate: 0.167
  cured ham: 0.333
  cookies: 1.000
  milk: 1.000
  watermelon: 0.000
Average clustering: 0.5476190476190477
Transitivity: 0.375

Exercise

Compute the local clustering coefficient of “chocolate” by hand from the drawing of G_foods. Then check your answer with nx.clustering.

Solution

“chocolate” has four neighbours (bread, cookies, milk and watermelon), so there are \(4 \cdot 3 / 2 = 6\) neighbour pairs. Only cookies and milk are connected, so \(C_{\text{chocolate}} = 1/6 \approx 0.167\).

print(nx.clustering(G_foods, "chocolate"))

0.16666666666666666

What’s Next?

The next page ranks nodes by centrality.

Centrality in NetworkX

References

Newman, Mark. 2018. Networks. 2nd ed. Oxford University Press.

Watts, Duncan J., and Steven H. Strogatz. 1998. “Collective Dynamics of ’Small-World’ Networks.” Nature 393 (6684): 440–42. https://doi.org/10.1038/30918.