Repositori institucional URV
Belongs to TFM:SerieGeneralMESIIA collection
TITLE:
Improving stability of GNNExplainer inlarge citation network datasets - TFM:1426
Handle:
https://hdl.handle.net/20.500.11797/TFM1426
Student:
Cabezas Rodriguez, José Joaquin
Language:
en
Title in original language:
Improving stability of GNNExplainer inlarge citation network datasets
Title in different languages:
Improving stability of GNNExplainer inlarge citation network datasets
Keywords:
graph neural networks, explainability, machine learning
Subject:
Aprenentatge automàtic
Abstract:
Graph Neural Networks (GNNs) is a Machine Learning framework that brings neural networks to graph and relational data. It is of special relevance for areas like social network analysis, biological sciences, chemistry, smart transportation systems and many others, where data can be thought of as a network. Explaining why a GNN made a decision is a challenge, due to the black-box nature of neural networks, but it is crucial when applying it to decision-making processes that affects the life of many. In this work we review the current state of the art and analyze the most well-known method for explaining GNNs, GNNExplainer. We find that its application to academic citations datasets present issues due to the variability of the explanations and we propose a modification for improving stability of the results and interpretability of the graphical explanation. In particular, we propose the use of an adjusted coefficient computed beforehand for every explanation instead of a fixed parameter. We find that our proposal improves the stability by more than 10 \% in experiments using Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) with two citation networks datasets (Cora and Pubmed).
Project director:
Duch Gavaldà, Jordi
Department:
Enginyeria Informàtica i Matemàtiques
Education area(s):
Enginyeria de la Seguretat Informàtica i Intel·ligència Artificial
Entity:
Universitat Rovira i Virgili (URV)
Work's public defense date:
2021-02-08
Academic year:
2020-2021
Confidenciality:
No
Subject areas:
Industrial Engineering
APS:
NO
Creation date in repository:
2023-07-011
Access Rights:
info:eu-repo/semantics/openAccess
Coverage:
NO
Type:
info:eu-repo/semantics/masterThesis
Títol:
Improving stability of GNNExplainer inlarge citation network datasets
Contributor:
Duch Gavaldà, Jordi
Subject:
Enginyeria informàtica
Industrial Engineering
Ingeniería informática
Aprenentatge automàtic
Date:
2021-02-08
Language:
en
Format:
Universitat Rovira i Virgili (URV)
Creator:
Cabezas Rodriguez, José Joaquin
Rights:
info:eu-repo/semantics/openAccess
Search your record at:
Available files
File
Description
Format
Memòria
Memory
application/pdf
View/Open
Show entire record
Go back
All objects of this collection
Information
© 2011 Universitat Rovira i Virgili
Legal
Accessibility
Contact
