On the Potential of Network-Based Features for Fraud Detection

TL;DR

This paper investigates the use of personalized PageRank (PPR) to enhance fraud detection models by capturing social dynamics between financial accounts, showing that PPR improves predictive accuracy and provides valuable, stable features.

Contribution

It introduces the application of PPR for fraud detection, demonstrating its effectiveness in improving model performance over traditional features.

Findings

PPR integration improves fraud detection accuracy.

PPR features have high importance scores.

Feature distributions remain stable across datasets.

Abstract

Online transaction fraud presents substantial challenges to businesses and consumers, risking significant financial losses. Conventional rule-based systems struggle to keep pace with evolving fraud tactics, leading to high false positive rates and missed detections. Machine learning techniques offer a promising solution by leveraging historical data to identify fraudulent patterns. This article explores using the personalised PageRank (PPR) algorithm to capture the social dynamics of fraud by analysing relationships between financial accounts. The primary objective is to compare the performance of traditional features with the addition of PPR in fraud detection models. Results indicate that integrating PPR enhances the model's predictive power, surpassing the baseline model. Additionally, the PPR feature provides unique and valuable information, evidenced by its high feature importance score. Feature stability analysis confirms consistent feature distributions across training and test datasets.