Short Ticketing Detection Framework Analysis Report
This addresses fraud detection for railway operators, but appears incremental as it combines existing algorithms without major methodological breakthroughs.
The study tackled short ticketing fraud detection in railway systems using an unsupervised multi-expert framework, resulting in the identification of five distinct fraud patterns across 30 high-risk stations.
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.