Graph Neural Networks for Natural Language Processing: A Survey

Графовые нейронные сети для обработки естественного языка: обзор
Jian Pei, Lingfei Wu, Xiaojie Guo, Hanning Gao, Shucheng Li, Bo Long, Yu Chen, Kai Shen
2023-01-25

GNN taxonomyGraph Neural NetworksNLP tasksNatural Language Processinggraph-based deep learning
Deep learning has become the dominant approach in addressing various tasks in Natural Language Processing (NLP). Although text inputs are typically represented as a sequence of tokens, there is a rich variety of NLP problems that can be best expressed with a graph structure. As a result, there is a surge of interest in developing new deep learning techniques on graphs for a large number of NLP tasks. In this survey, we present a comprehensive overview on Graph Neural Networks (GNNs) for Natural Language Processing. We propose a new taxonomy of GNNs for NLP, which systematically organizes existing research of GNNs for NLP
1
It provides a comprehensive overview of existing applications and research on GNNs for Natural Language Processing.
2
The paper introduces a new taxonomy that systematically organizes GNN research in NLP.
3
The survey examines Graph Neural Networks as deep learning techniques for NLP tasks naturally represented by graph structures.
4
The survey highlights growing interest in graph-based deep learning because many NLP problems extend beyond sequential token representations.

Graph Neural Networks applied to Natural Language Processing tasks

The taxonomy, techniques, and applications of Graph Neural Networks for representing and solving diverse Natural Language Processing problems with graph structures

Publication Details
Publication Date
2023-01-25
Journal
Publisher
ISSN
Cited by
283
Access Type
Author Information
Authors
Jian Pei
Lingfei Wu
Xiaojie Guo
Hanning Gao
Shucheng Li
Bo Long
Yu Chen
Kai Shen
Explore further
Open the scid.ai AI chat with a ready-made request: it will find papers on a similar topic and help build a literature review.
Find similar papers in the chat →
Make a presentation
100%