Fraud Fingerprinting Pipeline for B2B Transaction Risk Scoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Organizations face challenges in detecting fraud or suspicious activity across multiple interactions without replacing their existing data processing frameworks, as they receive increasing amounts of data from various entities, making them vulnerable to internal and external attacks, particularly in B2B payment fraud.
Innovation Solution
A cloud-based fraud detection system that ingests raw data, processes it into structured objects, applies classification and fingerprinting processes, enriches fingerprints with external data, and determines verdicts using a verdict engine, allowing integration with existing systems and leveraging historical data for immediate risk assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If organizations use existing data processing frameworks to handle increasing data flow, then operational continuity is maintained, but fraud detection capability deteriorates
Solution Approach 1:
The fraud detection system is segmented into independent modular components including data ingestion engine, classification service, fingerprinting process, enrichment engine, and verdict engine. Each module performs a specific function and can be independently deployed, maintained, and scaled, allowing integration with existing frameworks without requiring complete system replacement.
Solution Approach 2:
The patent introduces an intermediary fraud detection layer that sits between existing data processing frameworks and the core business logic. This intermediary layer ingests data from existing systems, performs fraud analysis through multiple processing stages, and returns verdicts without disrupting the underlying framework operations.
2Measurement precision
If comprehensive data verification is implemented across all interactions, then fraud detection accuracy is improved, but processing time increases
Solution Approach 1:
The system performs preliminary fingerprinting and classification on incoming data before full verification. The fingerprinting process creates unique identifiers and initial classifications that enable rapid matching against known fraudulent patterns, allowing most transactions to be processed quickly without requiring complete comprehensive analysis.
Solution Approach 2:
The fraud detection system operates continuously in parallel with existing data processing workflows. Multiple verification stages run concurrently rather than sequentially, and the system maintains continuous learning from new data patterns without interrupting the flow of legitimate transactions.
3Reliability
If multiple data sources are integrated for correlation analysis, then fraud detection comprehensiveness is improved, but system complexity increases
Solution Approach 1:
The data ingestion engine is designed with universal interfaces that can accept and process data from multiple heterogeneous sources including transaction systems, customer databases, external fraud databases, and third-party services. The enrichment engine provides multi-functional capabilities to correlate data across different sources using standardized processing routines.
Solution Approach 2:
The system creates standardized copies and representations of data from different sources in a unified format. The fingerprinting process generates standardized identifiers that represent complex multi-source data in a compact, comparable form, enabling correlation analysis without requiring direct integration of all underlying systems.
Data Source
AI summary
Methods and systems, including computer programs encoded on computer storage media, for detecting fraudulent or suspicious transactions, e.g., in a business-to-business context. One of the methods includes: receiving, at an ingestion engine, raw data from a plurality of heterogeneous data sources; using the ingestion engine to produce structured object data, the structured object data derived from the raw data; applying a classification service to produce a partial fingerprint from the structured object data wherein the partial fingerprint classifies the structured object data; applying a fingerprinting process to the partial fingerprint to produce a full fingerprint for the structured object data; applying an enrichment engine to add data to the full fingerprint to produce an enriched fingerprint for the structured object data; and applying a verdict engine to determine a verdict on the object based at least in part on the enriched fingerprint for the structured object data.