PU GNN: Chargeback Fraud Detection in P2E MMORPGs via Graph Attention Networks with Imbalanced PU Labels
This addresses fraud prevention for game operators in blockchain-based P2E games, but it is incremental as it adapts existing techniques like GNNs and PU learning to a new domain.
The paper tackles chargeback fraud detection in play-to-earn MMORPGs, where in-game goods linked to cryptocurrencies cannot be restored after fraud, and proposes PU GNN, a method using graph attention networks with PU loss and modified GraphSMOTE to handle imbalanced labels, achieving superior performance on three real-world datasets.
The recent advent of play-to-earn (P2E) systems in massively multiplayer online role-playing games (MMORPGs) has made in-game goods interchangeable with real-world values more than ever before. The goods in the P2E MMORPGs can be directly exchanged with cryptocurrencies such as Bitcoin, Ethereum, or Klaytn via blockchain networks. Unlike traditional in-game goods, once they had been written to the blockchains, P2E goods cannot be restored by the game operation teams even with chargeback fraud such as payment fraud, cancellation, or refund. To tackle the problem, we propose a novel chargeback fraud prediction method, PU GNN, which leverages graph attention networks with PU loss to capture both the players' in-game behavior with P2E token transaction patterns. With the adoption of modified GraphSMOTE, the proposed model handles the imbalanced distribution of labels in chargeback fraud datasets. The conducted experiments on three real-world P2E MMORPG datasets demonstrate that PU GNN achieves superior performances over previously suggested methods.