Determining the optimal number of GAT and GCN layers for node classification in graph neural networks

Noor, Humaira and Islam, Niful and Hossain Mukta, Md Saddam and Nur Shazwani, Kamarudin and Khan Raiaan, Mohaimenul Azam and Azam, Sami (2023) Determining the optimal number of GAT and GCN layers for node classification in graph neural networks. In: 8th International Conference on Software Engineering and Computer Systems, ICSECS 2023 , 25-27 August 2023 , Penang. pp. 111-116. (192961). ISBN 979-835031093-1

	Pdf Determining the optimal number of GAT and GCN.pdf Restricted to Repository staff only Download (491kB) | Request a copy
Preview	Pdf Determining the optimal number of GAT and GCN layers for node classification in graph neural networks_ABS.pdf Download (139kB) | Preview

Abstract

Node classification in complex networks plays an important role including social network analysis and recommendation systems. Some graph neural networks such as Graph Convolutional Networks (GCN) and Graph Attention Networks (GAT) have emerged as effective approaches for achieving high-performance classification in such tasks. However, constructing a graph neural network architecture is challenging particularly due to the complex task of determining the optimal number of layers. This study presents a mathematical formula for determining the optimal number of GCN and GAT hidden layers. The experiment was conducted on ten benchmark datasets, evaluating performance metrices such as accuracy, precision, recall, F1-score, and MCC for identifying the best estimation of number of hidden layers. According to the experimental findings, the number of GAT and GCN layers selected has a substantial impact on classification accuracy. Studies show that adding extra layers after the optimum number of layers has a negative or no impact on the classification performance. Our proposed approximation technique may provide valuable insights for enhancing efficiency and accuracy of the Graph Neural Network algorithms.

Item Type:	Conference or Workshop Item (Lecture)
Additional Information:	Indexed by Scopus
Uncontrolled Keywords:	GAT; GCN; Graph; Neural network; Node classification
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software T Technology > T Technology (General) T Technology > TA Engineering (General). Civil engineering (General)
Faculty/Division:	Faculty of Computing
Depositing User:	Mr Muhamad Firdaus Janih@Jaini
Date Deposited:	16 Apr 2024 04:15
Last Modified:	16 Apr 2024 04:15
URI:	http://umpir.ump.edu.my/id/eprint/40357
Download Statistic:	View Download Statistics

Actions (login required)

View Item