无监督多专家模型识别铁路短票欺诈，定位30个高风险站点
Short Ticketing Detection Framework Analysis Report
- 用四种算法融合检测异常购票行为
- 发现五类短票欺诈模式，可挽回运输损失
- 适合交通反欺诈团队和系统安全研究者
本报告全面分析了一种用于铁路系统中检测短票欺诈的无监督多专家机器学习框架。研究提出了一套A/B/C/D站分类体系，成功识别出30个高风险站点中的可疑行为模式。该框架整合了四种互补算法：孤立森林、局部离群因子、一类支持向量机和马氏距离。关键发现包括识别出五种不同类型的短票欺诈模式，并展现出在交通运输系统中挽回短票损失的潜力。
原文摘要 · Abstract (English)
This report presents a comprehensive analysis of an unsupervised multi-expert machine learning framework for detecting short ticketing fraud in railway systems. The study introduces an A/B/C/D station classification system that successfully identifies suspicious patterns across 30 high-risk stations. The framework employs four complementary algorithms: Isolation Forest, Local Outlier Factor, One-Class SVM, and Mahalanobis Distance. Key findings include the identification of five distinct short ticketing patterns and potential for short ticketing recovery in transportation systems.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。