发表机构
Central South University; University of Technology Sydney, Australian Artificial Intelligence Institute（中南大学; 悉尼科技大学，澳大利亚人工智能研究所）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究图欺诈检测问题，提出PREF-Gate框架，结合无标签图上下文与标签衍生邻域证据，通过验证门控选择专家或概率混合，在多数据集实验中取得较好结果，提供可审计决策管道。
AI 中文摘要
关系欺诈检测可利用无标签图上下文和标签衍生邻域证据，但这两个信息源遵循不同有效性条件。当查询节点自身标签或任何验证或测试标签进入邻域风险构建时，邻域风险无效。本文将此问题表述为出处受限关系证据使用，并提出PREF-Gate，一个具有两个固定专家和有限验证门的可审计决策框架。上下文专家使用无标签的属性、一跳均值、特征残差和度描述符。证据专家添加自排除、仅训练标签的邻域风险和经验贝叶斯摘要。测试推理前，门选择专家或三种预指定概率混合之一并固定决策阈值。在亚马逊、YelpChi和TFinance数据集上，使用五种相同分层分割和14种相同协议方法，PREF-Gate获得的平均AUPRC值分别为0.9085、0.8104和0.8913。该框架将有竞争力的排名性能与明确的标签出处合同、有限选择策略、失败核算和审查预算评估相结合，为图欺诈检测提供了一个可审计的基于知识的决策管道。
英文摘要
Relational fraud detection can exploit both label-free graph context and label-derived neighborhood evidence, but these two information sources obey different validity conditions. In particular, neighborhood risk becomes invalid when a queried node's own label, or any validation or test label, enters its construction. We formulate this issue as provenance-constrained relational evidence use and present PREF-Gate, an auditable decision framework with two fixed experts and a finite validation gate. The context expert uses attributes, one-hop means, feature residuals, and degree descriptors without labels. The evidence expert adds self-excluded, training-label-only neighborhood risk and empirical-Bayes summaries that expose support, uncertainty, availability, and shrinkage. Before test inference, the gate selects either expert or one of three pre-specified probability mixtures and fixes the decision threshold. On Amazon, YelpChi, and TFinance, using five identical stratified splits and 14 same-protocol methods, PREF-Gate obtains mean AUPRC values of 0.9085, 0.8104, and 0.8913. It selects the label-free expert on all Amazon and YelpChi splits and an evidence mixture on all TFinance splits. Thus, the main result is conditional rather than universal: label-derived relational evidence is useful only where held-out validation supports it. The framework couples competitive ranking performance with an explicit label-provenance contract, finite selection policy, failure accounting, and review-budget evaluation, providing an auditable knowledge-based decision pipeline for graph fraud detection.
