Abstract:Recent years have witnessed the significant success of applying graph neural networks (GNNs) in learning effective node representations for classification. However, current GNNs are mostly built under the balanced data-splitting, which is inconsistent with many real-world networks where the number of training nodes can be extremely imbalanced among the classes. Thus, directly utilizing current GNNs on imbalanced data would generate coarse representations of nodes in minority classes and ultimately compromise the classification performance. This therefore portends the importance of developing effective GNNs for handling imbalanced graph data. In this work, we propose a novel Distance-wise Prototypical Graph Neural Network (DPGNN), which proposes a class prototype-driven training to balance the training loss between majority and minority classes and then leverages distance metric learning to differentiate the contributions of different dimensions of representations and fully encode the relative position of each node to each class prototype. Moreover, we design a new imbalanced label propagation mechanism to derive extra supervision from unlabeled nodes and employ self-supervised learning to smooth representations of adjacent nodes while separating inter-class prototypes. Comprehensive node classification experiments and parameter analysis on multiple networks are conducted and the proposed DPGNN almost always significantly outperforms all other baselines, which demonstrates its effectiveness in imbalanced node classification. The implementation of DPGNN is available at \url{<a class="link-external link-https" href="https://github.com/YuWVandy/DPGNN" rel="external noopener nofollow">this https URL</a>}.

Entropy and Gravitation Based Dynamic Radius Nearest Neighbor Classification for Imbalanced Problem

Gravitational Fixed Radius Nearest Neighbor for Imbalanced Problem

Multi-View Learning with Gravitational Nearest Neighbor Classifier

Under-bagging Nearest Neighbors for Imbalanced Classification

Classifying Unbalanced Pattern Groups by Training Neural Network

Improving nearest neighbor classification with simulated gravitational collapse

Doubly Approximate Nearest Neighbor Classification

Gravitation Balanced Multiple Kernel Learning for Imbalanced Classification

Fuzzy Nearest Neighbor Partitioning Neural Network for Classification

k-nearest neighbor classification method for class-imbalanced problem

A Gegenbauer Neural Network with Regularized Weights Direct Determination for Classification

On the gravitation‐based classification: A novel algorithm using equilibrium points for enhanced learning and dimensionality reduction

Adaptive Ensemble Method Based on Spatial Characteristics for Classifying Imbalanced Data

Adaptive ensemble of classifiers with regularization for imbalanced data classification

A Local Mean Representation-based K-Nearest Neighbor Classifier

Adaptive Three-way KNN Classifier Using Density-based Granular Balls

Multi-matrices Entropy Discriminant Ensemble Learning for Imbalanced Problem

Large-margin Nearest Neighbor Classifiers Via Sample Weight Learning.

Distance-wise Prototypical Graph Neural Network in Node Imbalance Classification

The Extended Nearest Neighbor Classification

A Modified Nearest Neighbor Classification Approach Based on Class-Wise Local Information