Persona2vec: a flexible multi-role representations learning framework for graphs

Jisung Yoon; Kai-Cheng Yang; Woo-Sung Jung; Yong-Yeol Ahn

doi:10.7717/peerj-cs.439

Persona2vec: a flexible multi-role representations learning framework for graphs

PeerJ Comput Sci. 2021 Mar 30:7:e439. doi: 10.7717/peerj-cs.439. eCollection 2021.

Authors

Jisung Yoon^{1

2}, Kai-Cheng Yang², Woo-Sung Jung^{1

3

4}, Yong-Yeol Ahn^{2

5

6}

Affiliations

¹ Department of Industrial and Management Engineering, Pohang University of Science and Technology, Pohang, Republic of Korea.
² Center for Complex Networks and Systems Research, Luddy School of Informatics, Computing, and Engineering, Indiana University, Bloomington, IN, USA.
³ Department of Physics, Pohang University of Science and Technology, Pohang, Republic of Korea.
⁴ Asia Pacific Center for Theoretical Physics, Pohang, Republic of Korea.
⁵ Connection Science, Massachusetts Institute of Technology, Cambridge, MA, USA.
⁶ Network Science Institute, Indiana University, Bloomington, IN, USA.

Abstract

Graph embedding techniques, which learn low-dimensional representations of a graph, are achieving state-of-the-art performance in many graph mining tasks. Most existing embedding algorithms assign a single vector to each node, implicitly assuming that a single representation is enough to capture all characteristics of the node. However, across many domains, it is common to observe pervasively overlapping community structure, where most nodes belong to multiple communities, playing different roles depending on the contexts. Here, we propose persona2vec, a graph embedding framework that efficiently learns multiple representations of nodes based on their structural contexts. Using link prediction-based evaluation, we show that our framework is significantly faster than the existing state-of-the-art model while achieving better performance.

Keywords: Graph embedding; Link prediction; Overlapping community; Social context; Social network analysis.

Grants and funding

This work is supported by the Air Force Office of Scientific Research under award number FA9550-191-0391. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.