Updated on 2022-08-08 GMT+08:00

Louvain

Overview

Louvain is a modularity-based community detection algorithm with high efficiency and effect. It detects hierarchical community structures and aims to maximize the modularity of the entire community network.

Application Scenarios

This algorithm applies to scenarios such as community mining and hierarchical clustering.

Parameter Description

Table 1 Louvain algorithm parameters

Parameter

Mandatory

Description

Type

Value Range

Default Value

convergence

No

Convergence

Double

A real number between 0 and 1 (excluding 0 and 1)

0.00001

max_iterations

No

Maximum iterations

Int

1-2,000

100

weight

No

Weight of an edge

String

Empty or null character string

  • Empty: The default weight and distance are 1.
  • Character string: The attribute of the corresponding edge is the weight. When the edge does not have corresponding attribute, the weight is 1 by default.
NOTE:

The weight of an edge must be greater than 0.

weight

Precautions

This algorithm generates only the final community result and does not save the hierarchical results.

Example

Set parameters coverage to 0.00001 and max_iterations to 100, the sub-graphs of different communities are displayed on the canvas. The color of a node varies with communities. The JSON result is displayed in the query result area.