GenCAT: Generating Attributed Graphs with Controlled Relationships between Classes, Attributes, and Topology

TL;DR

GenCAT is a novel attributed graph generator that allows precise control over relationships between labels, attributes, and topology, supporting scalable and high-quality synthetic graph creation for research purposes.

Contribution

It introduces the first generator supporting user-controlled relationships among labels, attributes, and topology, with scalable linear complexity.

Findings

GenCAT efficiently generates high-quality attributed graphs.

It supports flexible control over label-attribute relationships.

The generator scales linearly with the number of edges.

Abstract

Generating large synthetic attributed graphs with node labels is an important task to support various experimental studies for graph analysis methods. Existing graph generators fail to simultaneously simulate the relationships between labels, attributes, and topology which real-world graphs exhibit. Motivated by this limitation, we propose GenCAT, an attributed graph generator for controlling those relationships, which has the following advantages. (i) GenCAT generates graphs with user-specified node degrees and flexibly controls the relationship between nodes and labels by incorporating the connection proportion for each node to classes. (ii) Generated attribute values follow user-specified distributions, and users can flexibly control the correlation between the attributes and labels. (iii) Graph generation scales linearly to the number of edges. GenCAT is the first generator to support all three of these practical features. Through extensive experiments, we demonstrate that GenCAT can efficiently generate high-quality complex attributed graphs with user-controlled relationships between labels, attributes, and topology.