Akademik Veri Yönetim Sistemi
2025 Innovations in Intelligent Systems and Applications Conference, ASYU 2025, Bursa, Türkiye, 10 - 12 Eylül 2025, (Tam Metin Bildiri)
Blockchain-based loyalty platforms can enhance brand loyalty, boosting customer retention. Yet just like any software they remain vulnerable to fraud. We present a transparent fraud-detection framework for PointXchange, a blockchain-based customer loyalty platform that enables brands to collaboratively manage and exchange reward points in a decentralized environment. The study evaluates three gradient boosting machines and two neural networks on one public and one private dataset. To interpret our model, we apply SHAP, LIME, and LRP, in addition to introducing a surrogate-based fidelity assessment. The top performing predictors on two datasets achieve up to 0.89 and 0.93 F1-scores, respectively, while the top performing explanations yield a fidelity score of complete agreement. These results demonstrate the potential of explainable fraud detection in blockchain-based loyalty systems, contributing to both security and transparency by not just empirically evaluating existing approaches but tailoring them to a real-world system.