A Comprehensive Survey on Graph Neural Networks

By Z. Wu et al.
Published on Aug. 10, 2019
Read the original document by opening this link in a new tab.

Table of Contents

Abstract
Index Terms
I. I NTRODUCTION
II. B ACKGROUND & D EFINITION
III. C ATEGORIZATION AND FRAMEWORKS
IV. R ECURRENT G RAPH N EURAL N ETWORKS (REC GNN S)
V. T AXONOMY OF G RAPH N EURAL N ETWORKS (G NNS)
VI. G RAPH A UTOENCODERS (G AES)
VII. S PATIAL- T EMPORAL G RAPH N EURAL N ETWORKS (STGNN S)
VIII. C ONCLUSION

Summary

This document provides a comprehensive overview of graph neural networks (GNNs) in the fields of data mining and machine learning. It proposes a new taxonomy categorizing GNNs into recurrent GNNs, convolutional GNNs, graph autoencoders, and spatial-temporal GNNs. The document discusses the challenges posed by graph data and the emergence of deep learning approaches to address these challenges. It covers a wide range of topics including the applications, models, and resources related to GNNs. The paper also points towards future research directions in the rapidly evolving field of graph neural networks.
×
This is where the content will go.