A Survey of Network Representation Learning Methods for Link Prediction in Biological Network

Author(s): Jiajie Peng, Guilin Lu, Xuequn Shang*

Journal Name: Current Pharmaceutical Design

Volume 26 , Issue 26 , 2020

Become EABM
Become Reviewer

Abstract:

Background: Networks are powerful resources for describing complex systems. Link prediction is an important issue in network analysis and has important practical application value. Network representation learning has proven to be useful for network analysis, especially for link prediction tasks.

Objective: To review the application of network representation learning on link prediction in a biological network, we summarize recent methods for link prediction in a biological network and discuss the application and significance of network representation learning in link prediction task.

Method & Results: We first introduce the widely used link prediction algorithms, then briefly introduce the development of network representation learning methods, focusing on a few widely used methods, and their application in biological network link prediction. Existing studies demonstrate that using network representation learning to predict links in biological networks can achieve better performance. In the end, some possible future directions have been discussed.

Keywords: Biological network, link prediction, network analysis, network representation learning, algorithms, development.

Rights & PermissionsPrintExport Cite as

Article Details

VOLUME: 26
ISSUE: 26
Year: 2020
Page: [3076 - 3084]
Pages: 9
DOI: 10.2174/1381612826666200116145057
Price: $65

Article Metrics

PDF: 7