New SilIF Algorithm Enhances Unsupervised Fraud Detection

Researchers have developed SilIF, a novel enhancement to the Isolation Forest algorithm for unsupervised transaction fraud detection. SilIF incorporates a silhouette-based scoring layer, which analyzes the representation space created by the forest's trees to better identify anomalies. This augmentation demonstrated an improvement in AUC-PR by 0.0080 on average over standard Isolation Forest on a large transaction fraud dataset, though its effectiveness varied across different datasets. AI

IMPACT Introduces a tunable enhancement for unsupervised fraud detection, potentially improving accuracy in financial applications.

RANK_REASON The cluster contains a research paper detailing a new algorithm for unsupervised anomaly detection. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.LG →

• IEEE-CIS Fraud Detection benchmark
• Isolation Forest
• Venkatakrishnan Gopalakrishnan

• paper
• model release

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →