GATSMOTE: Improving Imbalanced Node Classification on Graphs via Attention and Homophily

Abstract

In recent decades, non-invasive neuroimaging techniques and graph theories have enabled a better understanding of the structural patterns of the human brain at a macroscopic level. As one of the most widely used non-invasive techniques, an electroencephalogram (EEG) may collect non-neuronal signals from “bad channels”. Automatically detecting these bad channels represents an imbalanced classification task; research on the topic is rather limited. Because the human brain can be naturally modeled as a complex graph network based on its structural and functional characteristics, we seek to extend previous imbalanced node classification techniques to the bad-channel detection task. We specifically propose a novel edge generator considering the prominent small-world organization of the human brain network. We leverage the attention mechanism to adaptively calculate the weighted edge connections between each node and its neighboring nodes. Moreover, we follow the homophily assumption in graph theory to add edges between similar nodes. Adding new edges between nodes sharing identical labels shortens the path length, thus facilitating low-cost information messaging.

Prediction and optimization of employee turnover intentions in enterprises based on unbalanced data

“…Therefore, Pei W [ 22 ] proposed the Synthetic Minority Oversampling Technique (SMOTE) based on the analysis of the proximity between sample points to synthesize the minority class; then, they [ 23 ] proposed the feature selection method to solve the problem of high-dimensional unbalanced datasets. By analyzing the intrinsic relationships existing between samples through non-random sampling, the original feature information is maintained in the sampling process, making the classification model less prone to overfitting and misclassification during the training process [ 24 , 25 ]. Cost-sensitive learning reduces the error of SVM in classifying minority classes by reducing the overall cost of misclassification and improves the classification of unbalanced data [ 26 ].…”

Imbalance Node Classification with Graph Neural Networks (GNN): A Study on a Twitter Dataset

“…In real world applications, we are faced with imbalanced data and GNNs fail to accurately predict the samples that belong to the minority class. Imbalanced class problems can be solved by modifying the GNN model to bias toward minority class, or by resampling that consists in altering the dataset by adjusting the number of instances for each class to achieve a balanced distribution [16], [14]. The most used approach is resampling because it can be integrated with any classifier [16], [14].…”

“…Imbalanced class problems can be solved by modifying the GNN model to bias toward minority class, or by resampling that consists in altering the dataset by adjusting the number of instances for each class to achieve a balanced distribution [16], [14]. The most used approach is resampling because it can be integrated with any classifier [16], [14]. Resampling can be achieved either by undersampling or oversampling.…”

“…To this purpose, we applied different graph neural networks (GCN [10], GraphSage [11] and GAT [12]) on our dataset for the downstream task of classifying nodes. Given the fact that the dataset is imbalanced (there are more active users than inactive users), models which claim to fix this problem [13], [14], [15], [16], [17] were also studied.…”

Network-Induced Soft Sets and Stock Market Applications

“…Graph theory is an area of discrete mathematics whose popularity has increased in recent years, not only regarding theoretical developments, but also regarding its applications for numerous aspects [39][40][41][42][43][44][45]. Graph theory can be easily used in solving daily life problems, and it has been employed quite frequently in more theoretical studies, data analyses, and artificial intelligence applications.…”