Automated Payment Transaction Review System
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current payment transaction anti-fraud systems are inefficient due to the high time and cost associated with manual review of transactions, which accounts for over 50% of overall expenditure, and struggle to accurately differentiate between legitimate and fraudulent transactions.
Innovation Solution
The implementation of a system that uses electronic signatures based on user attributes such as browser fingerprints, computer fingerprints, IP addresses, and typing patterns to authenticate and identify users, allowing for automated verification and reducing the need for manual review by correlating newly collected data with previously stored data to determine transaction legitimacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual review is used to evaluate transactions flagged for review, then transaction legitimacy can be assessed, but time consumption and operational costs increase significantly
Solution Approach 1:
The system enables automated self-assessment of transactions by implementing an automated review component that evaluates transactions previously flagged for manual review. This component uses established fraud detection rules and criteria to automatically determine whether to accept or reject transactions, eliminating the need for human intervention in routine cases and significantly reducing review time while maintaining consistent assessment quality.
Solution Approach 2:
The patent replaces the mechanical manual review process with an automated electronic system. The automated review component substitutes human analysts with algorithm-based evaluation that processes transactions electronically using predefined fraud detection rules, thereby eliminating the time-consuming manual assessment while preserving the ability to detect fraudulent patterns through systematic electronic analysis.
2Reliability
If manual review is used to evaluate transactions, then fraud detection accuracy can be maintained, but operational costs increase by over 50% of total anti-fraud expenditure
Solution Approach 1:
The automated review component performs self-assessment of transactions using established fraud detection rules, eliminating the need for expensive manual human review. The system automatically evaluates transactions against predefined criteria and makes acceptance or rejection decisions, thereby maintaining fraud detection accuracy while reducing operational costs associated with human labor to minimal levels.
Solution Approach 2:
The system creates and applies replicated fraud detection rules and criteria across multiple transactions automatically. By copying proven fraud detection logic into the automated review component, the system maintains the reliability and accuracy of manual review processes while enabling high-volume automated processing that dramatically reduces the per-transaction cost of fraud detection.
3Productivity
If automated rules are used to evaluate transactions, then processing speed increases, but false positives and false negatives increase
Solution Approach 1:
The system dynamically adjusts the stringency and application of fraud detection rules based on transaction characteristics and risk profiles. The automated review component can modify evaluation criteria in real-time, applying more stringent rules to high-risk transactions while using streamlined rules for low-risk cases, thereby maintaining high processing speed while minimizing false positives and false negatives through adaptive rule application.
Solution Approach 2:
The system incorporates feedback mechanisms where outcomes of automated reviews are continuously analyzed to refine and adjust fraud detection rules. By monitoring false positives and false negatives from automated processing, the system learns from errors and optimizes rule parameters, progressively improving evaluation accuracy while maintaining high processing speeds through automated operations.
Data Source
AI summary
An authentication system for electronic transactions occurring over a computerized network is provided that includes: (a) a transaction authentication system interface operatively coupled to the computerized network for receiving proposed transactions information over the computerized network; and (b) a transaction authentication tool, operatively coupled to the transaction authentication system interface, where the tool includes: an automated anti-fraud analysis component for analyzing transactions information from a plurality of proposed transactions and automatically rejecting a first subset of proposed transactions according to automatic rejection criteria and automatically assigning a second subset of transactions for additional review; and an automated acceptance component for analyzing the second subset of transactions and accepting a third subset of transactions from the second subset based upon automatic accepting criteria.