Automated Fraud Detection Rule Generation via Machine Learning

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Existing rule-based systems for fraud detection in financial institutions are inefficient and inaccurate due to outdated rules, high dependency on human expertise, and resource-intensive maintenance processes.

Innovation Solution

A system and method for programmatically automating rule creation through automated feature selection and rule performance analysis using machine learning (ML) algorithms and simulated annealing processes, reducing the need for manual intervention and enhancing rule accuracy and efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If manual rule creation and maintenance is performed by subject matter experts, then rule accuracy and domain knowledge integration are improved, but time consumption and operational complexity increase significantly

Engineering Contradiction:

Improverule accuracyVSAvoidtime consumption

Core Design Contradiction:

Measurement precisionVSLoss of time

Solution Approach 1:

The system enables automated rule creation where the fraud detection rules are generated automatically by the machine learning system based on historical data and patterns, eliminating the need for manual rule creation by subject matter experts while maintaining high accuracy through data-driven insights

Inventive Principle:

Principle #25Self-service

Solution Approach 2:

The manual mechanical process of rule creation and maintenance by experts is replaced with an automated machine learning system that uses algorithms to analyze data, identify patterns, and generate rules automatically, significantly reducing time consumption while maintaining or improving rule accuracy

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

2Device complexity

If traditional rule-based systems are used for fraud detection, then implementation simplicity is maintained, but adaptability to changing fraud patterns and environmental changes deteriorates

Engineering Contradiction:

Improvesystem simplicityVSAvoidadaptability to changing patterns

Core Design Contradiction:

Device complexityVSAdaptability or versatility

Solution Approach 1:

The system implements dynamic rule generation where rules are continuously updated and adapted based on new data and changing fraud patterns, allowing the system to evolve automatically without manual intervention while maintaining operational simplicity through automated processes

Inventive Principle:

Principle #15Dynamics

Solution Approach 2:

The system incorporates feedback mechanisms where detection results and new fraud patterns are continuously fed back into the machine learning model, which automatically adjusts and updates rules to adapt to changing environments and fraud tactics, maintaining both simplicity and adaptability

Inventive Principle:

Principle #23Feedback

3Reliability

If comprehensive rule sets are maintained to cover all fraud scenarios, then detection coverage is improved, but computational resource consumption and processing time increase

Engineering Contradiction:

Improvedetection coverageVSAvoidcomputational resource consumption

Core Design Contradiction:

ReliabilityVSUse of energy by moving object

Solution Approach 1:

The system extracts only the most relevant and high-impact rules from the comprehensive rule set based on their performance metrics and relevance to current fraud patterns, eliminating redundant rules and reducing computational overhead while maintaining detection coverage for critical fraud scenarios

Inventive Principle:

Principle #2Taking out (Extraction)

Solution Approach 2:

The system dynamically adjusts rule parameters and thresholds based on data analysis and performance feedback, optimizing the balance between detection coverage and computational efficiency by modifying rule sensitivity and priority levels without reducing overall detection capability

Inventive Principle:

Principle #35Parameter changes

4Measurement precision

If frequent rule updates are performed to maintain accuracy, then detection precision is improved, but system stability and operational consistency worsen

Engineering Contradiction:

Improvedetection precisionVSAvoidsystem stability

Core Design Contradiction:

Measurement precisionVSStability of the object's composition

Solution Approach 1:

The system performs rule updates at optimized intervals and triggers based on significant pattern changes rather than continuously, maintaining detection precision through periodic retraining and updates while ensuring system stability by avoiding unnecessary frequent changes that could disrupt operational consistency

Inventive Principle:

Principle #19Periodic action

Data Source

PatentUS20250029106A1Automated rule creation for fraud detection using a machine learning-driven computing framework

Publication Date: 2025.01.23 ACTIMIZE LIMITED

AI summary

A rule training system and methods are provided that are configured to generate machine learning rules for fraud detection based on an automated feature selection by a rule creation system. The system includes a processor and a computer readable medium operably coupled thereto, the computer readable medium comprising a plurality of instructions stored in association therewith that are accessible to, and executable by, the processor, to perform rule creation operations which include accessing a training dataset, performing a data preprocessing and a data sampling on the training dataset, obtaining a plurality of features from the processed and sampled dataset using a feature engineering operation, selecting a subset of features from the plurality of features using simulated annealing operations, generating the plurality of detection rules using the subset of features, iteratively selecting from the plurality of detection rules, and evaluating a rule performance of each rule in those selected.