Merchant Fraud Risk Scoring Using AI Prediction Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing fraud and risk management systems for acquirer banks are inadequate in detecting complex merchant fraud due to reliance on simple rule-based methodologies, failing to provide comprehensive insights, leading to significant financial losses, regulatory risks, and legal liabilities.
Innovation Solution
A computer-implemented method and system using artificial intelligence-based risk prediction models to generate fraud risk scores for merchants, including health, terminal, chargeback, and activity risk scores, based on transaction data, facilitating enhanced fraud detection and risk management.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If simple rule-based methodologies are used for merchant fraud detection, then the system complexity is reduced and ease of operation is improved, but the measurement precision and reliability of fraud detection deteriorate
Solution Approach 1:
The patent replaces simple rule-based mechanical detection systems with AI/ML-based intelligent prediction models. The system uses machine learning algorithms to analyze transaction patterns, merchant behavior, and risk indicators, substituting rigid mechanical rules with adaptive intelligent systems that can learn from historical data and improve detection accuracy over time.
Solution Approach 2:
The patent transforms the detection approach by changing from binary rule-based parameters (yes/no matches) to continuous probability scores generated by ML models. The system outputs risk probabilities and fraud scores that provide nuanced measurement precision, allowing for gradient-based risk assessment rather than threshold-based detection.
2Reliability
If AI-based risk prediction models are implemented, then the measurement precision and reliability of fraud detection are improved, but the device complexity increases
Solution Approach 1:
The patent segments the risk assessment system into multiple independent prediction models, each specialized for a specific risk type (e.g., card fraud, account takeover, merchant fraud, chargeback detection). This modular architecture allows the system to manage complexity by dividing the overall reliability challenge into manageable specialized components that can be developed, tested, and maintained independently.
Solution Approach 2:
The patent introduces an intermediary layer between raw transaction data and final risk decisions. The ML prediction models act as intermediaries that process complex data patterns and translate them into interpretable risk scores. This intermediary layer simplifies the decision-making process for acquirers while maintaining high reliability through sophisticated underlying analysis.
3Reliability
If comprehensive fraud detection capabilities are implemented, then the reliability of risk management is improved, but the loss of time in processing transactions increases
Solution Approach 1:
The patent applies preliminary action by pre-training ML models on historical fraud data and pre-configuring risk thresholds before actual transactions occur. The system performs preliminary risk assessment calculations in advance, allowing real-time transaction processing to benefit from pre-computed risk profiles and decision trees, thereby reducing processing time while maintaining comprehensive detection capabilities.
Solution Approach 2:
The patent implements partial action by selectively applying comprehensive AI analysis only to transactions that trigger risk thresholds or exhibit suspicious patterns. Low-risk transactions proceed through simplified processing paths, while high-risk transactions receive comprehensive ML analysis. This partial application of excessive action ensures reliability where needed without unnecessarily delaying routine transactions.
Data Source
AI summary
Methods and server systems for computing fraud risk scores for various merchants associated with an acquirer described herein. The method performed by a server system includes accessing merchant-related transaction data including merchant-related transaction indicators associated with a merchant from a transaction database. Method includes generating a merchant-related transaction features based on the merchant-related indicators. Method includes generating via risk prediction models, for a payment transaction with the merchant, merchant health and compliance risk scores, merchant terminal risk scores, merchant chargeback risk scores, and merchant activity risk scores based on the merchant-related transaction features. Method includes facilitating transmission of a notification message to an acquirer server associated with the merchant. The notification message includes the merchant health and compliance risk scores, the merchant terminal risk scores, the merchant chargeback risk scores, and the merchant activity risk scores.