PubMed-Temporal: A dynamic graph dataset with node-level features

Code to build and reproduce the temporal split for the PubMed/Planetoid graph dataset.

If you use this dataset in your research, please consider citing the paper that introduced it:

Passos, N.A.R.A., Carlini, E., Trani, S. (2024). Deep Community Detection in Attributed Temporal Graphs: Experimental Evaluation of Current Approaches. In Proceedings of the 3rd Graph Neural Networking Workshop 2024 (GNNet '24). Association for Computing Machinery, New York, NY, USA, 1–6.

Description

FIrst citation occurs from a paper published in 1967 to another published in 1964.

Load dataset

PyTorch Geometric

from pubmed_temporal import Planetoid
# from torch_geometric.datasets import Planetoid  # pytorch_geometric#9982

dataset = Planetoid(root=".", name="pubmed", split="temporal")
data = dataset[0]
print(data)

Data(x=[19717, 500], edge_index=[2, 88648], y=[19717], time=[88648],
     train_mask=[88648], val_mask=[88648], test_mask=[88648])

The number of edges is doubled in the undirected graph from PyTorch Geometric.

NetworkX

import networkx as nx

G = nx.read_graphml("pubmed/temporal/graph/pubmed-temporal.graphml")
print(G)

DiGraph with 19717 nodes and 44335 edges

The directed graph contains 11 bidirectional edges from co-citing papers.

Build dataset

The temporal split and edge masks for the train, validation, and test splits are already included in this repository.

In order to build it completely from scratch (requires pubmed-id), run:

python build_dataset.py --workers 1

To build the dataset, the following steps are taken, aside from obtaining the required data from PubMed:

1. Download original PubMed graph dataset.
2. Build NetworkX object from dataset.
3. Obtain Planetoid node index map.
4. Relabel nodes to match Planetoid's index map.
5. Add weight vectors x.
6. Add classes y.
7. Add time steps time.
8. Verify if dataset matches Planetoid's.
9. Save data with edge time steps starting from zero.

Extras

To plot the figures and table displayed above:

python extra/build_extra.py

Requires the extra requirements: matplotlib and tabulate.

References

• Query-driven Active Surveying for Collective Classification (2012). Namata et al., Workshop on Mining and Learning with Graphs (MLG), Edinburgh, Scotland, UK, 2012.

• Revisiting Semi-Supervised Learning with Graph Embeddings (2016). Yang et al., Proceedings of the 33rd International Conference on Machine Learning (ICML), New York, NY, USA, 2016.