ML Prompt-Based Data Connection Detection for Fraud Alerts
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Manual identification of data connections between customer datasets for fraud detection and anti-money laundering is time-consuming and inefficient, often requiring significant analyst time and increasing the risk of missed matches.
Innovation Solution
An automated system using machine learning models, such as large language models, generates connection analysis prompts to identify links between datasets, including customer and fraud/money laundering datasets, by applying prompts to machine learning models to produce alerts when connections are found.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual identification of data connections is used, then accuracy of match identification can be maintained, but time consumption and analyst effort increase significantly
Solution Approach 1:
The patent introduces an intermediary system comprising a processor and machine learning models that act as a mediator between raw customer data and fraud detection alerts. This intermediary automatically identifies data connections between customer datasets and fraud/money laundering datasets, eliminating the need for manual analyst intervention while maintaining high accuracy through sophisticated matching algorithms.
Solution Approach 2:
The patent replaces the mechanical manual process of analyst review with an automated electronic system using machine learning models. The system processes data connections through computational algorithms that can identify matches between datasets without human intervention, dramatically reducing time consumption while preserving detection accuracy.
2Productivity
If automated solutions are implemented, then time efficiency improves, but ability to identify complex data connections deteriorates
Solution Approach 1:
The patent employs machine learning models that can dynamically adjust their analysis parameters and approaches based on the complexity of data connections. The system processes various data types and connection patterns using adaptable algorithms that maintain high identification accuracy while operating automatically, thus preserving the ability to identify complex connections while improving time efficiency.
3Reliability
If manual analysis of large amounts of data is performed, then thorough investigation can be achieved, but analyst workload and resource requirements increase
Solution Approach 1:
The patent implements a self-service automated system that independently performs data connection identification without requiring analyst resources. The machine learning models autonomously process customer datasets, compare them against fraud and money laundering datasets, and generate alerts for potential matches, thereby maintaining thorough investigation quality while eliminating analyst workload.
Data Source
AI summary
A system and method for identifying data connections may include a computing device; a memory; and a processor, the processor configured to: generate a connection analysis prompt from one or more data items of a first dataset for identifying one or more data items of a second dataset; and apply said connection analysis prompt to a machine learning model to produce an output from the machine learning model of whether said one or more data items of the first dataset are connected to said one or more data items of the second dataset; and when said one or more data items of the first dataset have one or more connections to said one or more data items of a second dataset, to produce an alert.