Graph ML's annotated bibliography, with code

What it does

This repository is a curated index of graph classification research, split into four strands—matrix factorization, spectral and statistical fingerprints, deep learning, and graph kernels—where each listed paper is paired with a reference implementation. It functions as a living literature review that links theory to code, and it points to an external repository for benchmark datasets.

The interesting bit

The maintainer has built a whole constellation of these “awesome” paper-and-code lists—covering community detection, fraud detection, gradient boosting, and more—turning scattered subfields into reproducible reading lists.

Key highlights

• Four methodological chapters span classical kernels to modern deep graph networks
• Every entry ties a paper to its reference implementation
• Links to external benchmark datasets for standardized comparisons
• Released under CC0, so the index itself is public-domain reference material
• 4,800+ GitHub stars suggest it is a common starting point for graph ML surveys

Caveats

• The README is minimal; it does not assess implementation quality, maintenance status, or ease of use
• This is an index, not a framework: expect to hunt down and run individual paper repos rather than a unified pipeline

Verdict

Researchers and students mapping the graph classification landscape should start here; engineers needing a single, maintained library should keep scrolling.