Automated Decisioning Workflow for Fraud Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current digital fraud and abuse detection technologies lack accuracy and real-time responsiveness, failing to effectively detect new threats and adapt to evolving digital threats, leading to insufficient protection for service providers and users.
Innovation Solution
A method and system that utilize machine learning-based threat scores and automated decisioning workflows to detect and mitigate digital threats by configuring a testing group, computing an insult rate, and reconfiguring thresholds based on performance data, enabling real-time detection and adaptation to new threats.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If existing technology implementations are used to detect digital fraud and digital abuse, then some detection capability is provided, but the accuracy and real-time responsiveness are insufficient
Solution Approach 1:
The system dynamically adjusts detection thresholds and model parameters in real-time based on incoming data patterns and threat evolution. The automated decisioning workflow continuously reconfigures machine learning models without manual intervention, enabling the system to adapt its detection sensitivity dynamically while maintaining real-time operational reliability.
Solution Approach 2:
The system implements continuous feedback loops where detection results, false positives, and emerging threat patterns are fed back into the machine learning models. This feedback mechanism allows the system to self-correct and improve detection accuracy over time while maintaining real-time responsiveness through automated model retraining and threshold adjustment.
2Adaptability or versatility
If existing technology implementations are used, then some detection capability is provided, but the ability to detect new and never-before-encountered digital threats is lacking
Solution Approach 1:
The system performs self-service through automated machine learning model evolution and reconfiguration. The automated decisioning workflow independently detects new threat patterns, retrains models, and adjusts detection parameters without human intervention. This self-service capability enables continuous adaptation to emerging threats while maintaining high levels of automation.
Solution Approach 2:
The system prepares for new threats by continuously pre-training machine learning models on emerging threat patterns and maintaining multiple model versions. The automated workflow pre-configures detection parameters and thresholds based on anticipated threat evolution, enabling rapid response to novel digital threats before they fully manifest.
3Productivity
If machine learning-based threat score thresholds are used for automated decisioning, then detection automation is achieved, but false positives and negatives occur
Solution Approach 1:
The system dynamically changes detection parameters, thresholds, and model configurations based on real-time performance metrics and incoming data patterns. The automated decisioning workflow adjusts threat score thresholds and model parameters to optimize the balance between detection efficiency and accuracy, reducing false positives and negatives while maintaining high productivity.
Data Source
AI summary
A system and method for generating an insult rate and reconfiguring an automated decisioning workflow includes configuring a testing group based on sampling from online events having an adverse disposal decision computed by an automated decisioning workflow computer that is configured with machine learning-based threat score thresholds that, if satisfied, causes a computation of a disallow decision or a block decision for a given online event; evaluating a performance and collecting performance data of distinct members of the testing group over a testing period; computing an insult rate for the testing group based on the performance data; computing an insult rate equilibrium for the automated decisioning workflow computer based on the performance data; evaluating the insult rate against the insult rate equilibrium; and reconfiguring adverse decisioning thresholds based on the evaluation of the insult rate of the testing group against the insult rate equilibrium for the automated decisioning workflow computer.