A study on graph neural networks

This report investigates various Graph Neural Network (GNN) models and its performance and stability. GNNs have gained popularity in recent years because they are able to handle graph data structures, which are a common way to represent complex relationships between entities in many real-world appli...

Full description

Bibliographic Details
Main Author: Choo, Patricia Yu Wei
Other Authors: Tay Wee Peng
Format: Final Year Project (FYP)
Language:English
Published: Nanyang Technological University 2023
Subjects:
Online Access:https://hdl.handle.net/10356/167731
_version_ 1826117933268467712
author Choo, Patricia Yu Wei
author2 Tay Wee Peng
author_facet Tay Wee Peng
Choo, Patricia Yu Wei
author_sort Choo, Patricia Yu Wei
collection NTU
description This report investigates various Graph Neural Network (GNN) models and its performance and stability. GNNs have gained popularity in recent years because they are able to handle graph data structures, which are a common way to represent complex relationships between entities in many real-world applications. This project focuses on node classification problems and is tested on public benchmark datasets. The paper discusses the possible improvement in performance and stability using Bootstrapped Graph Latents (BGRL) and compare it to bootstrapping neural networks.
first_indexed 2024-10-01T04:35:28Z
format Final Year Project (FYP)
id ntu-10356/167731
institution Nanyang Technological University
language English
last_indexed 2024-10-01T04:35:28Z
publishDate 2023
publisher Nanyang Technological University
record_format dspace
spelling ntu-10356/1677312023-07-07T19:35:06Z A study on graph neural networks Choo, Patricia Yu Wei Tay Wee Peng School of Electrical and Electronic Engineering wptay@ntu.edu.sg Engineering::Computer science and engineering Engineering::Electrical and electronic engineering This report investigates various Graph Neural Network (GNN) models and its performance and stability. GNNs have gained popularity in recent years because they are able to handle graph data structures, which are a common way to represent complex relationships between entities in many real-world applications. This project focuses on node classification problems and is tested on public benchmark datasets. The paper discusses the possible improvement in performance and stability using Bootstrapped Graph Latents (BGRL) and compare it to bootstrapping neural networks. Bachelor of Engineering (Information Engineering and Media) 2023-06-03T13:36:58Z 2023-06-03T13:36:58Z 2023 Final Year Project (FYP) Choo, P. Y. W. (2023). A study on graph neural networks. Final Year Project (FYP), Nanyang Technological University, Singapore. https://hdl.handle.net/10356/167731 https://hdl.handle.net/10356/167731 en A3017-221 application/pdf Nanyang Technological University
spellingShingle Engineering::Computer science and engineering
Engineering::Electrical and electronic engineering
Choo, Patricia Yu Wei
A study on graph neural networks
title A study on graph neural networks
title_full A study on graph neural networks
title_fullStr A study on graph neural networks
title_full_unstemmed A study on graph neural networks
title_short A study on graph neural networks
title_sort study on graph neural networks
topic Engineering::Computer science and engineering
Engineering::Electrical and electronic engineering
url https://hdl.handle.net/10356/167731
work_keys_str_mv AT choopatriciayuwei astudyongraphneuralnetworks
AT choopatriciayuwei studyongraphneuralnetworks