Community discovery in networks with deep sparse filtering. (September 2018)

• Record Type: Journal Article
• Authors: Xie, Yu; Gong, Maoguo; Wang, Shanfeng; Yu, Bin
• Is Part Of: Pattern recognition. Volume 81(2018:Sep.)
• Page Start: 50
• Page End: 59
• Publication Date: 2018-09
• Subjects: Community discovery -- Sparse filtering -- Deep learning -- Network representation; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2018.03.026
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: A new method for network community discovery based on deep sparse filtering is first presented. We make full use of the network topology and choose four matrix forms to represent a network graph respectively. A proper constraint is put forward for the purpose of making the proposed algorithm learn more efficient features for community discovery. Abstract: In the past decade, network community discovery has attracted great attention from quite a few researchers, and community structure is one of the most significant properties in complex networks. This paper presents a novel method for network community discovery based on deep sparse filtering. The features of the network are extracted by sparse filtering, an unsupervised deep learning algorithm, from an efficient representation of the network. Consequently, extracted features are employed to partition the network. Experiment results on both synthetic and real-world network datasets indicate that the proposed algorithm especially based on S ⌀ rensen–Dice's similarity matrix representation of the network is efficient and it outperforms several state-of-art algorithms in discovering community structure.