Are heterophily-specific gnns and homophily metrics really effective? evaluation pitfalls and new benchmarks

S Luan, Q Lu, C Hua, X Wang, J Zhu… - arxiv preprint arxiv …, 2024 - arxiv.org
Over the past decade, Graph Neural Networks (GNNs) have achieved great success on
machine learning tasks with relational data. However, recent studies have found that …

Diverse joint nonnegative matrix tri-factorization for attributed graph clustering

A Mohammadi, SA Seyedi, FA Tab… - Applied Soft …, 2024 - Elsevier
Cluster analysis of attributed graphs is a demanding and challenging task in the analysis of
network-structured data. It involves learning node representation by leveraging both node …

On performance discrepancies across local homophily levels in graph neural networks

D Loveland, J Zhu, M Heimann, B Fish… - Learning on Graphs …, 2024 - proceedings.mlr.press
Abstract Graph Neural Network (GNN) research has highlighted a relationship between high
homophily (ie, the tendency of nodes of the same class to connect) and strong predictive …

SkyMap: a generative graph model for GNN benchmarking

A Wassington, R Higueras, S Abadal - Frontiers in Artificial …, 2024 - frontiersin.org
Graph Neural Networks (GNNs) have gained considerable attention in recent years. Despite
the surge in innovative GNN architecture designs, research heavily relies on the same 5-10 …

Integrating Fuzzy Clustering and Graph Convolution Network to Accurately Identify Clusters From Attributed Graph

Y Yang, G Li, D Li, J Zhang, P Hu… - IEEE Transactions on …, 2024 - ieeexplore.ieee.org
Attributed graph clustering is of significance for an in-depth understanding of the intrinsic
organization of complex networks. Recently, owing to the powerful learning capability of …

Efficient Dynamic Attributed Graph Generation

F Li, X Wang, D Cheng, C Chen, Y Zhang… - arxiv preprint arxiv …, 2024 - arxiv.org
Data generation is a fundamental research problem in data management due to its diverse
use cases, ranging from testing database engines to data-specific applications. However …

[HTML][HTML] From attributes to communities: a novel approach in social network generation

MÇ Uludağlı, K Oğuz - PeerJ Computer Science, 2024 - peerj.com
Generating networks with attributes would be useful in computer game development by
enabling dynamic social interactions, adaptive storylines, realistic economic systems …

Hyperbolic Benchmarking Unveils Network Topology-Feature Relationship in GNN Performance

R Aliakbarisani, R Jankowski, M Serrano… - arxiv preprint arxiv …, 2024 - arxiv.org
Graph Neural Networks (GNNs) have excelled in predicting graph properties in various
applications ranging from identifying trends in social networks to drug discovery and …