FincGAN: A Gan Framework of Imbalanced Node Classification on Heterogeneous Graph Neural Network
Authors: Hung-Chun Hsu*, Ting-Le Lin*, Bo-Jun Wu, Ming-Yi Hong, Che Lin, Chih-Yu Wang
In Proceedings of IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), 2024
TL;DR
FincGAN is a node generation framework designed to address the issue of node class imbalance, aiming to prevent Graph Neural Network (GNN) models trained on such imbalanced graph data from exhibiting a bias toward predicting the majority class in node classification tasks.
FincGAN leverages a Generative Adversarial Network (GAN) to produce synthetic nodes, which are then integrated into the original graph to alleviate the class imbalance. However, generating synthetic nodes that closely resemble real nodes remains a challenging problem. For further details, please refer to our paper!