# cdlib.algorithms.symmnmf¶

symmnmf(g_original: object, dimensions: int = 32, iterations: int = 200, rho: float = 100.0, seed: int = 42) → cdlib.classes.node_clustering.NodeClustering

The procedure decomposed the second power od the normalized adjacency matrix with an ADMM based non-negative matrix factorization based technique. This results in a node embedding and each node is associated with an embedding factor in the created latent space.

Supported Graph Types

Undirected Directed Weighted
Yes No No
Parameters: g_original – a networkx/igraph object dimensions – Number of dimensions. Default is 32. iterations – Number of power iterations. Default is 200. rho – Regularization tuning parameter. Default is 100.0. seed – Random seed value. Default is 42. NodeClustering object
>>> from cdlib import algorithms
>>> import networkx as nx
>>> G = nx.karate_club_graph()
>>> coms = algorithms.symmnmf(G)


Kuang, Da, Chris Ding, and Haesun Park. “Symmetric nonnegative matrix factorization for graph clustering.” Proceedings of the 2012 SIAM international conference on data mining. Society for Industrial and Applied Mathematics, 2012.

Note