Real-time Fraud Detection via Non-Transaction Data Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for detecting malicious activity, particularly social engineering-based payment fraud, are inefficient due to the inability to process non-transaction data in real-time, leading to delayed transaction execution and potential misclassification of fraudulent transactions.
Innovation Solution
A computer-implemented method that integrates an analytics engine into existing payment systems to analyze non-transaction data in real-time, converting non-standardized data into a standardized format, and transmitting risk indicators to determine transaction risk, allowing for dynamic selection of data sources and improved fraud detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If non-transaction data is processed in real-time, then fraud detection accuracy is improved, but system complexity increases
Solution Approach 1:
The system segments the data processing function by introducing a separate analytics engine that operates independently from the transaction processing system. This analytics engine consumes non-transaction data from various sources, processes it through standardized formats, and outputs risk indicators that are then fed back into the transaction system for fraud detection decisions.
Solution Approach 2:
The patent introduces an integration layer as an intermediary component that sits between the transaction system and the analytics engine. This integration layer handles the communication protocols, data format conversions, and coordination between the different systems, thereby reducing the overall system complexity while enabling real-time processing.
2Productivity
If non-standardized data is converted to standardized format, then data processing efficiency is improved, but processing time increases
Solution Approach 1:
The system performs preliminary action by pre-defining standardized data formats and schemas before actual data processing occurs. The analytics engine is configured with expected data structures in advance, allowing for rapid validation and conversion of incoming non-standardized data without time-consuming on-the-fly format decisions.
Solution Approach 2:
The patent implements parameter changes by establishing a standardized data format specification that the analytics engine must adhere to. This standardization transforms variable, non-standardized input data into a consistent output format that can be efficiently processed and integrated with existing transaction systems, thereby improving processing efficiency despite the time required for conversion.
3Measurement precision
If data from multiple sources is consolidated, then detection accuracy is improved, but data volume increases
Solution Approach 1:
The system extracts only the relevant and necessary information from multiple non-transaction data sources. The analytics engine selectively processes data based on predefined criteria and risk indicators, filtering out irrelevant information while retaining only the data elements that contribute to accurate fraud detection. This extraction approach maintains high detection accuracy while managing data volume.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
This document describes a computer-implemented method that includes storing information in a standardized format about an organization's susceptibility to social engineering in a plurality of network-based, non-transitory storage devices having a collection of social engineering risk indicators stored thereon; importing, using an integration layer, non-standardized updated information about the organization from one or more data sources; converting, using a first analytics engine, the non-standardized updated information into the standardized format; and transmitting, via the integration layer, the standardized updated information for one or more organizations to a second analytics engine configured to determine a transaction risk indicator for a transaction.