Fraud-Preserving Dual-Horizon Temporal Heterogeneous Graph Contrastive Learning for Credit Card Fraud Detection
Academic Journal of Finance and Accounting · 05.11.2026 · 1 (5) · 73-86
DOI: 10.54691/d14yqy06 ↗Kurz gefasst
Fachartikel in der Zeitschrift Academic Journal of Finance and Accounting, online erschienen am 05.11.2026. Verfasst von Jia Kang und 1 weiteren Autor:innen. Die Arbeit nennt 28 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC 4.0).
Abstract (Originalsprache)
Learning from financial transaction graphs requires both recency sensitivity and retention of sparse, long-range behavioral evidence. We study a fraud-preserving adaptive augmentation (FPAA) mechanism in a dual-horizon heterogeneous graph encoder. Each transaction receives strictly earlier user-, merchant-, and device-linked messages, with a recent exponential-decay path and a long-memory path. Training combines focal classification, TH-GCL-style structurally and temporally perturbed positive pairs, and class-conditioned edge survival applied only to observed training labels. On a fully synthetic, reproducible 30,000-event stream (2.40% fraud; chronological 60/20/20 split), the proposed dual-horizon FPAA model achieves average precision 0.427 ± 0.038 across three training seeds, versus 0.421 ± 0.032 for uniform perturbations and 0.441 ± 0.032 for the dual-horizon supervised encoder. The short-memory-only supervised encoder achieves 0.368, indicating that retaining long-range history is consequential in this benchmark. A logged view-survival audit confirms that FPAA preserves 96.0% of fraud-linked training edges under structural augmentation, compared with 71.2% for uniform dropout. Across three independently generated synthetic streams with the same simulator design, FPAA mean AP is 0.423, against 0.417 for ordinary contrastive augmentation; a causally history-enriched LightGBM reaches 0.502. Hence, protecting fraud edges changes view integrity but provides a modest predictive gain, and historical information alone explains much of the initially observed graph advantage.
Zitieren
APA
Kang, J., & Adler, Z. (2026). Fraud-Preserving Dual-Horizon Temporal Heterogeneous Graph Contrastive Learning for Credit Card Fraud Detection. Academic Journal of Finance and Accounting, 1(5), 73-86. https://doi.org/10.54691/d14yqy06
Vancouver
Kang J, Adler Z. Fraud-Preserving Dual-Horizon Temporal Heterogeneous Graph Contrastive Learning for Credit Card Fraud Detection. Academic Journal of Finance and Accounting. 2026;1(5):73-86. doi:10.54691/d14yqy06
BibTeX
@article{Kang2026,
author = {Kang, Jia and Adler, Zain},
title = {Fraud-Preserving Dual-Horizon Temporal Heterogeneous Graph Contrastive Learning for Credit Card Fraud Detection},
journal = {Academic Journal of Finance and Accounting},
year = {2026},
volume = {1},
number = {5},
pages = {73--86},
doi = {10.54691/d14yqy06}
}
EUVIMED