Dual Node and Edge Fairness-Aware Graph Partition
This work addresses fairness in unsupervised user analysis for social networks, offering an incremental improvement by extending existing methods to consider edge-level biases.
The paper tackles the problem of fair graph partitioning in social networks by introducing edge balance to complement existing node balance, resulting in partitions that achieve both node and edge fairness while maintaining good utility.
Fair graph partition of social networks is a crucial step toward ensuring fair and non-discriminatory treatments in unsupervised user analysis. Current fair partition methods typically consider node balance, a notion pursuing a proportionally balanced number of nodes from all demographic groups, but ignore the bias induced by imbalanced edges in each cluster. To address this gap, we propose a notion edge balance to measure the proportion of edges connecting different demographic groups in clusters. We analyze the relations between node balance and edge balance, then with line graph transformations, we propose a co-embedding framework to learn dual node and edge fairness-aware representations for graph partition. We validate our framework through several social network datasets and observe balanced partition in terms of both nodes and edges along with good utility. Moreover, we demonstrate our fair partition can be used as pseudo labels to facilitate graph neural networks to behave fairly in node classification and link prediction tasks.