ExplainabilityAccountabilityBiascs.LG
Aiding Humans in Financial Fraud Decision Making: Toward an XAI-Visualization Framework
Angelos Chatzimparmpas, Evanthia Dimara
August 26, 2024
arXiv.org
4 citations
Key Takeaways
-
Humans remain the final decision-makers in fraud cases due to regulatory requirements, but they currently struggle to synthesize vast amounts of information from multiple sources (AI alerts, transaction histories, social media, and legal requirements), creating significant cognitive burden.
-
Current visual analytics tools are fragmented and incomplete, addressing only isolated aspects of fraud investigation (such as explaining individual AI alerts or showing transaction patterns) rather than supporting investigators through the entire decision-making process from data collection through final judgment.
-
A comprehensive visual analytics framework can reduce bias and human error by supporting investigators at all stages of investigation—data collection, information synthesis, and criteria iteration—while keeping human judgment central to final decisions rather than relying solely on automated AI systems.
-
Visual analytics can serve as a bridge between AI and human expertise, ensuring that AI tools inform rather than replace human decision-making in high-stakes financial fraud cases where regulatory compliance and fairness are critical concerns.
Abstract
AI prevails in financial fraud detection and decision making. Yet, due to concerns about biased automated decision making or profiling, regulations mandate that final decisions are made by humans. Financial fraud investigators face the challenge of manually synthesizing vast amounts of unstructured information, including AI alerts, transaction histories, social media insights, and governmental laws. Current Visual Analytics (VA) systems primarily support isolated aspects of this process, such as explaining binary AI alerts and visualizing transaction patterns, thus adding yet another layer of information to the overall complexity. In this work, we propose a framework where the VA system supports decision makers throughout all stages of financial fraud investigation, including data collection, information synthesis, and human criteria iteration. We illustrate how VA can claim a central role in AI-aided decision making, ensuring that human judgment remains in control while minimizing potential biases and labor-intensive tasks.