Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Enhancing Financial Fraud Detection with Human-in-the-Loop Feedback and Feedback Propagation

Prashank Kadam

TL;DR

The results show that HITL feedback improves model accuracy, with graph-based techniques benefiting the most, and a novel feedback propagation method is introduced that extends feedback across the dataset, further enhancing detection accuracy.

Abstract

Human-in-the-loop (HITL) feedback mechanisms can significantly enhance machine learning models, particularly in financial fraud detection, where fraud patterns change rapidly, and fraudulent nodes are sparse. Even small amounts of feedback from Subject Matter Experts (SMEs) can notably boost model performance. This paper examines the impact of HITL feedback on both traditional and advanced techniques using proprietary and publicly available datasets. Our results show that HITL feedback improves model accuracy, with graph-based techniques benefiting the most. We also introduce a novel feedback propagation method that extends feedback across the dataset, further enhancing detection accuracy. By leveraging human expertise, this approach addresses challenges related to evolving fraud patterns, data sparsity, and model interpretability, ultimately improving model robustness and streamlining the annotation process.

Enhancing Financial Fraud Detection with Human-in-the-Loop Feedback and Feedback Propagation

TL;DR

The results show that HITL feedback improves model accuracy, with graph-based techniques benefiting the most, and a novel feedback propagation method is introduced that extends feedback across the dataset, further enhancing detection accuracy.

Abstract

Human-in-the-loop (HITL) feedback mechanisms can significantly enhance machine learning models, particularly in financial fraud detection, where fraud patterns change rapidly, and fraudulent nodes are sparse. Even small amounts of feedback from Subject Matter Experts (SMEs) can notably boost model performance. This paper examines the impact of HITL feedback on both traditional and advanced techniques using proprietary and publicly available datasets. Our results show that HITL feedback improves model accuracy, with graph-based techniques benefiting the most. We also introduce a novel feedback propagation method that extends feedback across the dataset, further enhancing detection accuracy. By leveraging human expertise, this approach addresses challenges related to evolving fraud patterns, data sparsity, and model interpretability, ultimately improving model robustness and streamlining the annotation process.

Paper Structure

This paper contains 24 sections, 3 equations, 2 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Human-in-the-Loop (HITL) Feedback
  4. Financial Fraud Detection
  5. Method
  6. Manual Annotation
  7. Feedback Propagation
  8. Graph Construction
  9. Initialization
  10. Propagation
  11. Experiments
  12. Datasets
  13. PFFD
  14. Yelp Dataset
  15. Training and Evaluation
  16. ...and 9 more sections

Figures (2)

  • Figure 1: Transaction Graph - Here the red node has been manually annotated where the $isFraud$ score has been set to 100. You can see here how the score is discounted based on the edge weights and the node similarities and propogated further through the graph
  • Figure 2: AUC and Recall plots for Test Iterations of the PFFD dataset with and without Feedback Propagation