Name
Abstract | ||
|---|---|---|
ABSTRACTAbusive behavior in online retail websites and communities threatens the experience of regular community members. Such behavior often takes place within a complex, dynamic, and large-scale network of users interacting with items. Detecting abuse is challenging due to the scarcity of labeled abuse instances and complexity of combining temporal and network patterns while operating at a massive scale. Previous approaches to dynamic graph modeling either do not scale, do not effectively generalize from a few labeled instances, or compromise performance for scalability. Here we present BiDyn, a general method to detect abusive behavior in dynamic bipartite networks at scale, while generalizing from limited training labels. BiDyn develops an efficient hybrid RNN-GNN architecture trained via a novel stacked ensemble training scheme. We also propose a novel pre-training framework for dynamic graphs that helps to achieve superior performance at scale. Our approach outperforms recent large-scale dynamic graph baselines in an abuse classification task by up to 14% AUROC while requiring 10x less memory per training batch in both open and proprietary datasets. |
Year | DOI | Venue |
|---|---|---|
2021 | 10.1145/3447548.3467141 | Knowledge Discovery and Data Mining |
Keywords | DocType | Citations |
Fraud Detection, Anomaly Detection, Graph Neural Networks | Conference | 1 |
PageRank | References | Authors |
0.40 | 0 | 6 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Andrew Z Wang | 1 | 2 | 0.74 |
| rex ying | 2 | 787 | 23.66 |
| Pan Li | 3 | 41 | 11.95 |
| Nikhil S. Rao | 4 | 178 | 15.75 |
| Karthik Subbian | 5 | 1 | 0.40 |
| Jure Leskovec | 6 | 18769 | 886.50 |