Multitask Active Learning for Graph Anomaly Detection

Requirements

This code requires the following:

• Python==3.8
• PyTorch==2.0.1
• Numpy==1.24.4
• DGL==1.1.1+cu102

Usage

Take Cora dataset as an example:

python main.py  --dataset cora --strategy_ad medoids_spec_nent_diff --device 0 --alpha 1.25 --beta 0.5 --gamma 1 --cluster_num 24 --tau 0.95  --phi 1.25 

The hyperparameters for other datasets are reported as follows.

	Cora	Citeseer	BlogCatalog	Flickr	Amazon	YelpChi
$\tau$	0.95	0.90	0.98	0.98	0.999	0.80
$\alpha$	1.25	0.50	1.25	1.25	1.00	0.40
$\beta$	0.50	2.00	1.00	0.50	1.00	1.25
$\phi$	1.25	2.00	1.00	0.50	10	10
$m$	24	24	18	27	10	20

The Amazon and YelpChi datasets can be found from GADBench.