Community Detection via Autoencoder-Like Nonnegative Tensor Decomposition

Jiewen Guan, Bilian Chen*, Xin Huang

*Corresponding author for this work

Research output: Contribution to journalJournal articlepeer-review


Community detection aims at partitioning a network into several densely connected subgraphs. Recently, nonnegative matrix factorization (NMF) has been widely adopted in many successful community detection applications. However, most existing NMF-based community detection algorithms neglect the multihop network topology and the extreme sparsity of adjacency matrices. To resolve them, we propose a novel conception of adjacency tensor, which extends adjacency matrix to multihop cases. Then, we develop a novel tensor Tucker decomposition-based community detection method—autoencoder-like nonnegative tensor decomposition (ANTD), leveraging the constructed adjacency tensor. Distinct from simply applying tensor decomposition on the constructed adjacency tensor, which only works as a decoder, ANTD also introduces an encoder component to constitute an autoencoder-like architecture, which can further enhance the quality of the detected communities. We also develop an efficient alternative updating algorithm with convergence guarantee to optimize ANTD, and theoretically analyze the algorithm complexity. Moreover, we also study a graph regularized variant of ANTD. Extensive experiments on real-world benchmark networks by comparing 27 state-of-the-art methods, validate the effectiveness, efficiency, and robustness of our proposed methods.
Original languageEnglish
Pages (from-to)4179-4191
Number of pages13
JournalIEEE Transactions on Neural Networks and Learning Systems
Issue number3
Early online date28 Sept 2022
Publication statusPublished - Mar 2024

Scopus Subject Areas

  • Software
  • Computer Science Applications
  • Computer Networks and Communications
  • Artificial Intelligence

User-Defined Keywords

  • Community detection
  • graph clustering
  • nonnegative tensor decomposition
  • optimization


Dive into the research topics of 'Community Detection via Autoencoder-Like Nonnegative Tensor Decomposition'. Together they form a unique fingerprint.

Cite this