Modularity-based representation learning for networks

• 来源期刊: 中国物理B（英文版）
• 年，卷（期）: 2020,（12）
• 所属期刊栏目: INTERDISCIPLINARY PHYSICS AND RELATED AREAS OF SCIENCE AND TECHNOLOGY
• 页码范围: 663-670
• 页数: 8页
• 语种: 英文
• DOI: 10.1088/1674-1056/abbbec

摘要

Network embedding aims at learning low-dimensional representation of vertexes in a network and effectively pre-serving network structures.These representations can be used as features for many complex tasks on networks such as community detection and multi-label classification.Some classic methods based on the skip-gram model have been pro-posed to learn the representation of vertexes.However,these methods do not consider the global structure(i.e.,community structure)while sampling vertex sequences in network.To solve this problem,we suggest a novel sampling method which takes community information into consideration.It first samples dense vertex sequences by taking advantage of modularity function and then learns vertex representation by using the skip-gram model.Experimental results on the tasks of commu-nity detection and multi-label classification show that our method outperforms three state-of-the-art methods on learning the vertex representations in networks.