CODEWITHSUNDEEP
python-2.7graphnetworkx
ericmjl
ericmjl

Reputation: 14684

NetworkX - Finding the natural clusters of points on a graph

Still kinda new to NetworkX here, but I wanted to be able to query a NetworkX graph to find all nodes within a cluster of nodes. Here is an example graph I had generated: Network of Nodes

As you can see, there are clusters of nodes. Within each cluster, every node is connected to every other node. You can see that in a zoom-up of five such clusters below: Zoom in on 5 clusters

I'd like to be able to find out how I can extract out each individual cluster of nodes. Each node is named with a long name ("A/Vietnam/2009/8/431", for example), and I know how to fish out an individual node using NetworkX's functions, but I don't know how to get all connected nodes within that cluster. Language of preference is Python (2.7), and I have been using the NetworkX package in conjunction.

Thanks everybody!

Upvotes: 8

Views: 9094

Answers (2)

Vikram
Vikram

Reputation: 1775

import networkx as nx
from networkx.algorithms.components import connected_components

n = 10
G = nx.empty_graph(n)

# G is the networkx graph 
subgraphs = [G.subgraph(c) for c in connected_components(G)]

# n gives the number of subgraphs
n_subgraphs = len(subgraphs)

# you can now loop through all nodes in each subgraph
for i in range(n_subgraphs):
    print(f"Subgraph: {i} consists of {subgraphs[i].nodes()}")

Upvotes: 15

Billy Raseman
Billy Raseman

Reputation: 311

Update for Python 3.x:

import networkx as nx

# Identify groups of connected components (or subgraphs)
sub_graphs = nx.connected_components(G)

# Loop through all nodes in each subgraph and display them
while True:
    try:
        print(next(sub_graphs))
    except StopIteration:
        break

Upvotes: 2

Related Questions