Data that are best represented as a graph such as social, biological, communication, or transportation networks, and energy grids are ubiquitous in our world today. As more of such structured and semi-structured data is becoming available, the machine learning methods that can leverage the signal in these data are becoming more valuable, and the importance of being able to effectively mine and learn from such data is growing.
These graphs are typically multi-relational, dynamic, and large-scale. Understanding the different techniques applicable to graph data, dealing with their heterogeneity and applications of methods for information integration and alignment, handling dynamic and changing graphs, and addressing each of these issues at scale are some of the challenges in developing machine learning methods for graph data that appear in a variety of applications.
In this special issue, we aim to publish articles that help us better understand the principles, limitations, and applications of current graph-based machine learning methods, and to inspire research on new algorithms, techniques, and domain analysis for machine learning with graphs.
Lead guest editor
Shobeir Fakhraei, Information Science Institute, Univ. of Southern California
Guest editors
Austin Benson, Computer Science Department, Cornell University
Ciro Cattuto, ISI Foundation
Danai Koutra, Computer Science & Engineering, University of Michigan
Vagelis Papalexakis, Computer Science & Engineering, UC Riverside
Jiliang Tang, Computer Science & Engineering Dept., Michigan State University